The Pinot Controller is responsible for a number of things

• Controllers maintain the global metadata (e.g. configs and schemas) of the system with the help of Zookeeper which is used as the persistent metadata store.

• Controllers host Helix Controller and is responsible for managing other pinot components (brokers, servers, minions)

• They maintain the mapping of which servers are responsible for which segments. This mapping is used by the servers, to download the portion of the segments that they are responsible for. This mapping is also used by the broker to decide which servers to route the queries to.

• Controller has admin endpoints for viewing, creating, updating and deleting configs which help us manage and operate the cluster.

• Controllers also have endpoints for segment uploads which are used in offline data pushes. They are responsible for initializing realtime consumption and coordination of persisting the realtime segments into the segment store periodically.

• They undertake other management activities such as managing retention of segments, validations.

There can be multiple instances of Pinot controller for redundancy. If there are multiple controllers, Pinot expects that all of them are configured with the same back-end storage system so that they have a common view of the segments (e.g. NFS). Pinot can use other storage systems such as HDFS or ADLS.

Starting a Controller

Make sure you've setup Zookeeper. If you're using docker, make sure to pull the pinot docker image. To start a controller

docker run \
    --network=pinot-demo \
    --name pinot-controller \
    -p 9000:9000 \
    -d ${PINOT_IMAGE} StartController \
    -zkAddress pinot-zookeeper:2181

bin/pinot-admin.sh StartController \
  -zkAddress localhost:2181 \
  -clusterName PinotCluster \
  -controllerPort 9000

Pinot Minion is a new component which leverages the Helix Task Framework . It can be attached to an existing Pinot cluster and then execute tasks as provided by the controller. It's a generic and single place for running background jobs. They help offload computationally intensive tasks—such as adding indexes to segments and merging segments—from other components.

Starting Minion

// coming soon

A table is a logical abstraction to refer to a collection of related data. It consists of columns and rows (documents).

Data in Pinot tables is sharded into segments. A Pinot table is modeled as a Helix resource. Each segment of a table is modeled as a Helix Partition.

A table is typically associated with a schema, which is used to define the names, data types and other information of the columns of the table.

There are 3 types of a Pinot table

Note that the query does not know the existence of offline or realtime tables. It only specifies the table name in the query. For example, regardless of whether we have an offline table myTable_OFFLINE , or a realtime table myTable_REALTIME or a hybrid table containing both of these, the query will simply use mytable as select count(*) from myTable .

A table config file is used to define the table properties, such as name, type, indexing, routing, retention etc. It is written in JSON format, and stored in the property store in Zookeeper, along with the table schema.

Offline Table Config

Here's an example table config for an offline table

"OFFLINE": {
    "tableName": "pinotTable", 
    "tableType": "OFFLINE",
    "quota": {
      "maxQueriesPerSecond": 300, 
      "storage": "140G"
    }, 
    "routing": {
      "segmentPrunerType": "partition", 
      "instanceSelectorType": "replicaGroup"
    }, 
    "segmentsConfig": {
      "schemaName": "pinotTable", 
      "timeColumnName": "daysSinceEpoch", 
      "timeType": "DAYS",
      "replication": "3", 
      "retentionTimeUnit": "DAYS", 
      "retentionTimeValue": "365", 
      "segmentPushFrequency": "DAILY", 
      "segmentPushType": "APPEND" 
    }, 
    "tableIndexConfig": { 
      "invertedIndexColumns": ["foo", "bar", "moo"], 
      "createInvertedIndexDuringSegmentGeneration": false, 
      "sortedColumn": [],
      "bloomFilterColumns": [],
      "starTreeIndexConfigs": [],
      "noDictionaryColumns": [],
      "onHeapDictionaryColumns": [], 
      "varLengthDictionaryColumns": [],
      "loadMode": "MMAP", 
      "columnMinMaxValueGeneratorMode": null
    },  
    "tenants": {
      "broker": "myBrokerTenant", 
      "server": "myServerTenant"
    },
    "metadata": {
      "customConfigs": {
        "key": "value", 
        "key": "value"
      }
    }
  }
}

We will now discuss each section of the table config in detail.

Top level fields

Second level fields

quota

routing

segmentsConfig

tableIndexConfig

Realtime Table Config

Here's an example table config for a realtime table. All the fields from the offline table config are valid for the realtime table. Additionally, realtime tables use some extra fields.

"REALTIME": { 
    "tableName": "pinotTable", 
    "tableType": "REALTIME", 
    "segmentsConfig": {
      "schemaName": "pinotTable", 
      "timeColumnName": "daysSinceEpoch", 
      "timeType": "DAYS",
      "replicasPerPartition": "3", 
      "retentionTimeUnit": "DAYS", 
      "retentionTimeValue": "5", 
      "completionConfig": {
        "completionMode": "DOWNLOAD"
      } 
    }, 
    "tableIndexConfig": { 
      "invertedIndexColumns": ["foo", "bar", "moo"], 
      "sortedColumn": ["column1"],
      "noDictionaryColumns": ["metric1", "metric2"],
      "loadMode": "MMAP", 
      "aggregateMetrics": true, 
      "streamConfigs": {
        "realtime.segment.flush.threshold.size": "0", 
        "realtime.segment.flush.threshold.time": "24h", 
        "stream.kafka.broker.list": "XXXX", 
        "stream.kafka.consumer.factory.class.name": "XXXX", 
        "stream.kafka.consumer.prop.auto.offset.reset": "largest", 
        "stream.kafka.consumer.type": "XXXX", 
        "stream.kafka.decoder.class.name": "XXXX", 
        "stream.kafka.decoder.prop.schema.registry.rest.url": "XXXX", 
        "stream.kafka.decoder.prop.schema.registry.schema.name": "XXXX", 
        "stream.kafka.hlc.zk.connect.string": "XXXX", 
        "stream.kafka.topic.name": "XXXX", 
        "stream.kafka.zk.broker.url": "XXXX", 
        "streamType": "kafka"
      }  
    },  
    "tenants": {
      "broker": "myBrokerTenant", 
      "server": "myServerTenant", 
      "tagOverrideConfig": {}
    },
    "metadata": {
    }
}

We will now discuss the sections have some behavior differences for realtime tables.

segmentsConfig

replicasPerPartition The number of replicas per partition for the realtime stream

completionConfig Holds information related to realtime segment completion. There is just one field in this config as of now, which is the completionMode. The value of the completioMode decides how non-committers servers should replace the in-memory segment during realtime segment completion. By default, if the in memory segment in the non-winner server is equivalent to the committed segment, then the non-committer server builds and replaces the segment, else it download the segment from the controller.

Currently, the supported value for completionMode is

• DOWNLOAD: In certain scenarios, segment build can get very memory intensive. It might become desirable to enforce the non-committer servers to just download the segment from the controller, instead of building it again. Setting this completionMode ensures that the non-committer servers always download the segment.

  For more details on why this is needed, check out Completion Config

tableIndexConfig

sortedColumn Indicates the column which should be sorted when creating the realtime segment

aggregateMetrics Aggregate the realtime stream data as it is consumed, where applicable, in order to reduce segment sizes. We sum the metric column values of all rows that have the same value for dimension columns and create one row in a realtime segment for all such rows. This feature is only available on REALTIME tables. Only supported aggregation right now is sum. Also note that for this to work, all metrics should be listed in noDictionaryColumns and there should not be any multi value dimensions.

streamConfigs This section is where the bulk of settings specific to the realtime stream and consumption are found. This section is specific to tables of type REALTIME and is ignored if the table type is any other. See section on Ingesting Realtime Data for an overview of how realtime ingestion works.

Here is a minimal example of what the streamConfigs section may look like:

"streamConfigs" : {
  "realtime.segment.flush.threshold.size": "0",
  "realtime.segment.flush.threshold.time": "24h",
  "realtime.segment.flush.desired.size": "150M",
  "streamType": "kafka",
  "stream.kafka.consumer.type": "LowLevel",
  "stream.kafka.topic.name": "ClickStream",
  "stream.kafka.consumer.prop.auto.offset.reset" : "largest"
}

The streamType field is mandatory. In this case, it is set to kafka. StreamType of kafka is supported natively in Pinot. You can use default decoder classes and consumer factory classes. Pinot allows you to use other stream types with their own consumer factory and decoder classes (or, even other decoder and consumer factory for kafka if your installation formats kafka messages differently). See Pluggable Streams.

There are some configurations that are generic to all stream types, and others that are specific to stream types.

Configuration generic to all stream types

• realtime.segment.flush.threshold.size: Maximum number of rows to consume before persisting the consuming segment.

  Note that in the example above, it is set to 0. In this case, Pinot automatically computes the row limit using the value of realtime.segment.flush.desired.size described below. If the consumer type is HighLevel, then this value will be the maximum per consuming segment. If the consumer type is LowLevel then this value will be divided across all consumers being hosted on any one pinot-server.

  Default is 5000000.

• realtime.segment.flush.threshold.time: Maximum elapsed time after which a consuming segment should be persisted.

  The value can be set as a human readable string, such as "1d", "4h30m", etc. This value should be set such that it is not below the retention of messages in the underlying stream, but is not so long that it may cause the server to run out of memory.

  Default is "6h"

• realtime.segment.flush.desired.size: Desired size of the completed segments.

  This setting is supported only if consumer type is set to LowLevel. This value can be set as a human readable string such as "150M", or "1.1G", etc. This value is used when realtime.segment.flush.threshold.size is set to 0. Pinot learns and then estimates the number of rows that need to be consumed so that the persisted segment is approximately of this size. The learning phase starts by setting the number of rows to 100,000 (can be changed with the setting realtime.segment.flush.autotune.initialRows). and increasing to reach the desired segment size. Segment size may go over the desired size significantly during the learning phase. Pinot corrects the estimation as it goes along, so it is not guaranteed that the resulting completed segments are of the exact size as configured. You should set this value to optimize the performance of queries (i.e. neither too small nor too large)

  Default is "200M"

• realtime.segment.flush.autotune.initialRows: Initial number of rows for learning.

  This value is used only if realtime.segment.flush.threshold.size is set o 0 and the consumer type is LowLevel. See realtime.segment.flush.desired.size above.

  Default is "100K"

Configuration specific to stream types

All of these configuration items have the prefix stream.<streamType>. In the example above, the prefix is stream.kafka.

Important ones to note here are:

• stream.kafka.consumer.type: This should have a value of LowLevel (recommended) or HighLevel.

• stream.kafka.topic.name: Name of the topic from which to consume.

• stream.kafka.consumer.prop.auto.offset.reset: Indicates where to start consumption from in the stream.

  If the consumer type is LowLevel, This configuration is used only when the table is first provisioned. In HighLevel consumer type, it will also be used when new servers are rolled in, or when existing servers are replaced with new ones. You can specify values such as smallest or largest, or even 3d if your stream supports it. If you specify largest, the consumption starts from the most recent events in the data stream. This is the recommended way to create a new table. If you specify smallest then the consumption starts from the earliest event available in the data stream.

All the configurations that are prefixed with the streamType are expected to be used by the underlying stream. So, you can set any of the configurations described in the Kafka configuraton page can be set using the prefix stream.kafka and Kafka should pay attention to it.

More options are explained in the Pluggable Streams section.

tenants

Similar to the offline table, this section defines the server and broker tenant used for this table. More details about tenant can be found in Tenant.

tagOverrideConfig

A tagOverrideConfig can be added under the tenants section for realtime tables, to override tags for consuming and completed segments. For example:

  "broker": "brokerTenantName",
  "server": "serverTenantName",
  "tagOverrideConfig" : {
    "realtimeConsuming" : "serverTenantName_REALTIME"
    "realtimeCompleted" : "serverTenantName_OFFLINE"
  }
}

In the above example, the consuming segments will still be assigned to serverTenantName_REALTIME hosts, but once they are completed, the segments will be moved to serverTeantnName_OFFLINE. It is possible to specify the full name of any tag in this section (so, for example, you could decide that completed segments for this table should be in pinot servers tagged as allTables_COMPLETED). Refer to Moving Completed Segments section for learning more about this config.

Hybrid Table Config

A hybrid table is simply a table composed of 2 tables, 1 of type offline and 1 of type realtime, which share the same name. In such a table, offline segments may be pushed periodically (say, once a day). The retention on the offline table can be set to a high value (say, a few years) since segments are coming in on a periodic basis, whereas the retention on the realtime part can be small (say, a few days). Once an offline segment is pushed to cover a recent time period, the brokers automatically switch to using the offline table for segments in that time period, and use realtime table only to cover later segments for which offline data may not be available yet.

Here's a sample table config for a hybrid table.

"OFFLINE": {
    "tableName": "pinotTable", 
    "tableType": "OFFLINE", 
    "segmentsConfig": {
      ... 
    }, 
    "tableIndexConfig": { 
      ... 
    },  
    "tenants": {
      "broker": "myBrokerTenant", 
      "server": "myServerTenant"
    },
    "metadata": {
      ...
    }
  },
  "REALTIME": { 
    "tableName": "pinotTable", 
    "tableType": "REALTIME", 
    "segmentsConfig": {
      ...
    }, 
    "tableIndexConfig": { 
      ... 
      "streamConfigs": {
        ...
      },  
    },  
    "tenants": {
      "broker": "myBrokerTenant", 
      "server": "myServerTenant"
    },
    "metadata": {
    ...
    }
  }
}

Note that creating a hybrid table has to be done in 2 separate steps of creating an offline and realtime table individually.

Creating a table

Create a table config for your data, or see examples for all possible batch/streaming tables.

Prerequisites

1. Setup the cluster

2. Create broker and server tenants

Offline Table Creation

docker run \
    --network=pinot-demo \
    --name pinot-batch-table-creation \
    ${PINOT_IMAGE} AddTable \
    -schemaFile examples/batch/airlineStats/airlineStats_schema.json \
    -tableConfigFile examples/batch/airlineStats/airlineStats_offline_table_config.json \
    -controllerHost pinot-controller \
    -controllerPort 9000 \
    -exec

Executing command: AddTable -tableConfigFile examples/batch/airlineStats/airlineStats_offline_table_config.json -schemaFile examples/batch/airlineStats/airlineStats_schema.json -controllerHost pinot-controller -controllerPort 9000 -exec
Sending request: http://pinot-controller:9000/schemas to controller: a413b0013806, version: Unknown
{"status":"Table airlineStats_OFFLINE succesfully added"}

bin/pinot-admin.sh AddTable \
    -schemaFile examples/batch/airlineStats/airlineStats_schema.json \
    -tableConfigFile examples/batch/airlineStats/airlineStats_offline_table_config.json \
    -exec

# add schema
curl -F schemaName=@airlineStats_schema.json  localhost:9000/schemas

# add table
curl -i -X POST -H 'Content-Type: application/json' \
    -d @airlineStats_offline_table_config.json localhost:9000/tables

Check out the table config in the Rest API to make sure it was successfully uploaded.

Streaming Table Creation

docker run \
    --network pinot-demo --name=kafka \
    -e KAFKA_ZOOKEEPER_CONNECT=pinot-zookeeper:2181/kafka \
    -e KAFKA_BROKER_ID=0 \
    -e KAFKA_ADVERTISED_HOST_NAME=kafka \
    -d wurstmeister/kafka:latest

docker exec \
  -t kafka \
  /opt/kafka/bin/kafka-topics.sh \
  --zookeeper pinot-zookeeper:2181/kafka \
  --partitions=1 --replication-factor=1 \
  --create --topic flights-realtime

docker run \
    --network=pinot-demo \
    --name pinot-streaming-table-creation \
    ${PINOT_IMAGE} AddTable \
    -schemaFile examples/stream/airlineStats/airlineStats_schema.json \
    -tableConfigFile examples/docker/table-configs/airlineStats_realtime_table_config.json \
    -controllerHost pinot-controller \
    -controllerPort 9000 \
    -exec

Executing command: AddTable -tableConfigFile examples/docker/table-configs/airlineStats_realtime_table_config.json -schemaFile examples/stream/airlineStats/airlineStats_schema.json -controllerHost pinot-controller -controllerPort 9000 -exec
Sending request: http://pinot-controller:9000/schemas to controller: 8fbe601012f3, version: Unknown
{"status":"Table airlineStats_REALTIME succesfully added"}

bin/pinot-admin.sh StartZookeeper -zkPort 2191

bin/pinot-admin.sh  StartKafka -zkAddress=localhost:2191/kafka -port 19092

bin/pinot-admin.sh AddTable \
    -schemaFile examples/stream/airlineStats/airlineStats_schema.json \
    -tableConfigFile examples/stream/airlineStats/airlineStats_realtime_table_config.json \
    -exec

Check out the table config in the Rest API to make sure it was successfully uploaded.

Pinot is designed to deliver low latency queries on large datasets. In order to achieve this performance, Pinot stores data in a columnar format and adds additional indices to perform fast filtering, aggregation and group by.

Raw data is broken into small data shards and each shard is converted into a unit known as a segment. One or more segments together form a table, which is the logical container for querying Pinot using SQL/PQL.

Pinot Storage Model

Pinot uses a variety of terms which can refer to either abstractions that model the storage of data or infrastructure components that drive the functionality of the system.

Table

Similar to traditional databases, Pinot has the concept of a table—a logical abstraction to refer to a collection of related data. As is the case with RDBMS, a table is a construct that consists of columns and rows (documents) that are queried using SQL. A table is associated with a schema which defines the columns in a table as well as their data types.

As opposed to RDBMS schemas, multiple tables can be created in Pinot (real-time or batch) that inherit a single schema definition. Tables are independently configured for concerns such as indexing strategies, partitioning, tenants, data sources, and/or replication.

Segment

Pinot has a distributed systems architecture that scales horizontally. Pinot expects the size of a table to grow infinitely over time. In order to achieve this, all data needs to be distributed across multiple nodes. Pinot achieves this by breaking data into smaller chunks known as segments **(this is similar to shards/partitions in HA relational databases). Segments can also be seen as time-based partitions.

Tenant

In order to support multi-tenancy, Pinot has first class support for tenants. A table is associated with a tenant. This allows all tables belonging to a particular logical namespace to be grouped under a single tenant name and isolated from other tenants. This isolation between tenants provides different namespaces for applications and teams to prevent sharing tables or schemas. Development teams building applications will never have to operate an independent deployment of Pinot. An organization can operate a single cluster and scale it out as new tenants increase the overall volume of queries. Developers can manage their own schemas and tables without being impacted by any other tenant on a cluster.

By default, all tables belong to a default tenant named "default". The concept of tenants is very important, as it satisfies the architectural principle of a "database per service/application" without having to operate many independent data stores. Further, tenants will schedule resources so that segments (shards) are able to restrict a table's data to reside only on a specified set of nodes. Similar to the kind of isolation that is ubiquitously used in Linux containers, compute resources in Pinot can be scheduled to prevent resource contention between tenants.

Cluster

Logically, a cluster is simply a group of tenants. As with the classical definition of a cluster, it is also a grouping of a set of compute nodes. Typically, there is only one cluster per environment/data center. There is no needed to create multiple clusters since Pinot supports the concept of tenants. At LinkedIn, the largest Pinot cluster consists of 1000+ nodes distributed across a data center. The number of nodes in a cluster can be added in a way that will linearly increase performance and availability of queries. The number of nodes and the compute resources per node will reliably predict the QPS for a Pinot cluster, and as such, capacity planning can be easily achieved using SLAs that assert performance expectations for end-user applications.

Pinot Components

A Pinot cluster is comprised of multiple distributed system components. These components are useful to understand for operators that are monitoring system usage or are debugging an issue with a cluster deployment.

• Controller

• Server

• Broker

• Minion (optional)

The benefits of scale that make Pinot linearly scalable for an unbounded number of nodes is made possible through its integration with Apache Zookeeper and Apache Helix.

Pinot Controller

A controller is the core orchestrator that drives the consistency and routing in a Pinot cluster. Controllers are horizontally scaled as an independent component (container) and has visibility of the state of all other components in a cluster. The controller reacts and responds to state changes in the system and schedules the allocation of resources for tables, segments, or nodes. As mentioned earlier, Helix is embedded within the controller as an agent that is a participant responsible for observing and driving state changes that are subscribed to by other components.

In addition to cluster management, resource allocation, and scheduling, the controller is also the HTTP gateway for REST API administration of a Pinot deployment. A web-based query console is also provided for operators to quickly and easily run SQL/PQL queries.

Pinot Broker

A broker receives queries from a client and routes their execution to one or more Pinot servers before returning a consolidated response.

Pinot Server

Servers host segments (shards) that are scheduled and allocated across multiple nodes and routed on an assignment to a tenant (there is a single tenant by default). Servers are independent containers that scale horizontally and are notified by Helix through state changes driven by the controller. A server can either be a real-time server or an offline server.

A real-time and offline server have very different resource usage requirements, where real-time servers are continually consuming new messages from external systems (such as Kafka topics) that are ingested and allocated on segments of a tenant. Because of this, resource isolation can be used to prioritize high-throughput real-time data streams that are ingested and then made available for query through a broker.

Pinot Minion

Pinot minion is an optional component that can be used to run background tasks such as "purge" for GDPR (General Data Protection Regulation). As Pinot is an immutable aggregate store, records containing sensitive private data need to be purged on a request-by-request basis. Minion provides a solution for this purpose that complies with GDPR while optimizing Pinot segments and building additional indices that guarantees performance in the presence of the possibility of data deletion. One can also write a custom task that runs on a periodic basis. While it's possible to perform these tasks on the Pinot servers directly, having a separate process (Minion) lessens the overall degradation of query latency as segments are impacted by mutable writes.

This section is a reference for the definition of major components and logical abstractions used in Pinot. Please visit the Basic Concepts section to get a general overview that ties together all of the reference material in this section.

Operator reference

Developer reference

What is Pinot?

Pinot is a real-time distributed OLAP datastore, built to deliver scalable real-time analytics with low latency. It can ingest from batch data sources (such as Hadoop HDFS, Amazon S3, Azure ADLS, Google Cloud Storage) as well as stream data sources (such as Apache Kafka).

Pinot was built by engineers at LinkedIn and Uber and is designed to scale up and out with no upper bound. Performance always remains constant based on the size of your cluster and an expected query per second (QPS) threshold.

Get started

Our documentation is structured to let you quickly get to the content you need and is organized around the different concerns of users, operators, and developers. If you're new to Pinot and want to learn things by example, please take a look at our getting started section.

Starter guides

To start importing data into Pinot, check out our guides on batch import and stream ingestion based on our plugin architecture.

Query example

Pinot works very well for querying time series data with many dimensions and metrics over a vast unbounded space of records that scales linearly on a per node basis. Filters and aggregations are both easy and fast.

SELECT sum(clicks), sum(impressions) FROM AdAnalyticsTable
  WHERE 
       ((daysSinceEpoch >= 17849 AND daysSinceEpoch <= 17856)) AND 
       accountId IN (123456789)
  GROUP BY 
       daysSinceEpoch TOP 100

Pinot supports SQL for querying read-only data. Learn more about querying Pinot for time series data in our PQL (Pinot Query Language) guide.

Installation

Pinot may be deployed to and operated on a cloud provider or a local or virtual machine. You may get started either with a bare-metal installation or a Kubernetes one (either locally or in the cloud). To get immediately started with Pinot, check out these quick start guides for bootstrapping a Pinot cluster using Docker or Kubernetes.

Standalone mode

Cluster mode

Learn

For a high-level overview that explains how Pinot works, please take a look at our basic concepts section.

To understand the distributed systems architecture that explains Pinot's operating model, please take a look at our basic architecture section.

Overview

This section focuses on answering the most frequently asked questions for people exploring the newly evolving category of distributed OLAP engines. Pinot was created by authors at both Uber and LinkedIn and has been hardened and battle tested at the very highest of load and scale.

Is Pinot a data warehouse or a database?

While Pinot doesn't match the typical mold of a database product, it is best understood based on your role as either an analyst, data scientist, or application developer.

Enterprise business intelligence

For analysts and data scientists, Pinot is best viewed as a highly-scalable data platform for business intelligence. In this view, Pinot converges big data platforms with the traditional role of a data warehouse, making it a suitable replacement for analysis and reporting.

Enterprise application development

For application developers, Pinot is best viewed as an immutable aggregate store that sources events from streaming data sources, such as Kafka, and makes it available for query using SQL.

As is the case with a microservice architecture, data encapsulation ends up requiring each application to provision its own data store, as opposed to sharing one OLTP database for reads and writes. In this case, it becomes difficult to query the complete view of a domain because it becomes stored in many different databases. This is costly in terms of performance, since it requires joins across multiple microservices that expose their data over HTTP under a REST API. To prevent this, Pinot can be used to aggregate all of the data across a microservice architecture into one easily queryable view of the domain.

Pinot tenants prevent any possibility of sharing ownership of database tables across microservice teams. Developers can create their own query models of data from multiple systems of record depending on their use case and needs. As with all aggregate stores, query models are eventually consistent and immutable.

Companies using Pinot

Features

• A column-oriented database with various compression schemes such as Run Length, Fixed Bit Length

• Pluggable indexing technologies - Sorted Index, Bitmap Index, Inverted Index

• Ability to optimize query/execution plan based on query and segment metadata

• Near real time ingestion from streams and batch ingestion from Hadoop

• SQL-like language that supports selection, aggregation, filtering, group by, order by, distinct queries on data

• Support for multi-valued fields

• Horizontally scalable and fault-tolerant

When should I use it?

Pinot is designed to execute OLAP queries with low latency. It is suited in contexts where fast analytics, such as aggregations, are needed on immutable data, possibly, with real-time data ingestion.

User facing Analytics Products

Pinot was originally built at LinkedIn to power rich interactive real-time analytic applications such as Who Viewed Profile, Company Analytics, Talent Insights, and many more. UberEats Restaurant Manager is another example of a customer facing Analytics App. At LinkedIn, Pinot powers 50+ user-facing products, ingesting millions of events per second and serving 100k+ queries per second at millisecond latency.

Real-time Dashboard for Business Metrics

Pinot can be also be used to perform typical analytical operations such as slice and dice, drill down, roll up, and pivot on large scale multi-dimensional data. For instance, at LinkedIn, Pinot powers dashboards for thousands of business metrics. One can connect various BI tools such Superset, Tableau, or PowerBI to visualize data in Pinot.

Instructions to connect Pinot with Superset can found here.

Anomaly Detection

In addition to visualizing data in Pinot, one can run Machine Learning Algorithms to detect Anomalies on the data stored in Pinot. See ThirdEye for more information on how to use Pinot for Anomaly Detection and Root Cause Analysis.

This page will introduce you to the guiding principles behind the design of Apache Pinot. Here you will learn the distributed systems architecture that allows Pinot to scale the performance of queries linearly based on the number of nodes in a cluster. You'll also be introduced to the two different types of tables used to ingest and query data in offline (batch) or real-time (stream) mode.

Guiding design principles

Pinot was designed by engineers at LinkedIn and Uber to scale query performance based on the number of nodes in a cluster. As you add more nodes, query performance will always improve based on the expected query volume per second quota. To achieve horizontal scalability to an unbounded number of nodes and data storage, without performance degradation, the following guiding design principles were established.

• Highly available: Pinot is built to serve low latency analytical queries for customer facing applications. By design, there is no single point of failure in Pinot. The system continues to serve queries when a node goes down.

• Horizontally scalable: Ability to scale by adding new nodes as a workload changes.

• Latency vs Storage: Pinot is built to provide low latency even at high-throughput. Features such as segment assignment strategy, routing strategy, star-tree indexing were developed to achieve this.

• Immutable data: Pinot assumes that all data stored is immutable. For GDPR compliance, we provide an add-on solution for purging data while maintaining performance guarantees.

• Dynamic configuration changes: Operations such as adding new tables, expanding a cluster, ingesting data, modifying indexing config, and re-balancing must be performed without impacting query availability or performance.

Core components

As described in the concepts, Pinot has multiple distributed system components: Controller, Broker, Server, and Minion.

Pinot uses Apache Helix for cluster management. Helix is embedded as an agent within the different components and uses Apache Zookeeper for coordination and maintaining the overall cluster state and health.

Apache Helix and Zookeeper

All Pinot servers and brokers are managed by Helix. Helix is a generic cluster management framework to manage partitions and replicas in a distributed system. It's helpful to think of Helix as an event-driven discovery service with push and pull notifications that drives the state of a cluster to an ideal configuration. A finite-state machine maintains a contract of stateful operations that drives the health of the cluster towards its optimal configuration. Query load is optimized as Helix updates routing configurations between nodes based on where data is stored in the cluster.

Helix divides nodes into three logical components based on their responsibilities:

1. Participant: These are the nodes in the cluster that actually host the distributed storage resources.

2. Spectator: These nodes observe the current state of each participant and routes requests accordingly. Routers, for example, need to know the instance on which a partition is hosted and its state in order to route the request to the appropriate endpoint. Routing is continually being changed to optimize cluster performance as storage primitives are added and changed.

3. Controller: The controller observes and manages the state of participant nodes. The controller is responsible for coordinating all state transitions in the cluster and ensures that state constraints are satisfied while maintaining cluster stability.

Helix uses Zookeeper to maintain cluster state. Each component in a Pinot cluster takes a Zookeeper address as a startup parameter. The various components that are distributed in a Pinot cluster will watch Zookeeper notifications and issue updates via its embedded Helix-defined agent.

Helix agents use Zookeeper to store and update configurations, as well as for distributed coordination. Zookeeper stores the following information about the cluster:

Knowing the ZNode layout structure in Zookeeper for Helix agents in a cluster is useful for operations and/or troubleshooting cluster state and health.

Controller

Pinot's controller acts as the driver of the cluster's overall state and health. Because of its role as a Helix participant and spectator, which drives the state of other components, it is the first component that is typically started after Zookeeper. Two parameters are required for starting a controller: Zookeeper address and cluster name. The controller will automatically create a cluster via Helix if it does not yet exist.

Fault tolerance

To achieve fault tolerance, one can start multiple controllers (typically three) and one of them will act as a leader. If the leader crashes or dies, another leader is automatically elected. Leader election is achieved using Apache Helix. Having at-least one controller is required to perform any DDL equivalent operation on the cluster, such as adding a table or a segment.

The controller does not interfere with query execution. Query execution is not impacted even when all controllers nodes are offline. If all controller nodes are offline, the state of the cluster will stay as it was when the last leader went down. When a new leader comes online, a cluster resumes re-balancing activity and can accept new tables or segments.

Controller REST interface

The controller provides a REST interface to perform CRUD operations on all logical storage resources (servers, brokers, tables, and segments).

Broker

The responsibility of the broker is to route a given query to an appropriate server instance. A broker will collect and merge the responses from all servers into a final result and send it back to the requesting client. The broker provides HTTP endpoints that accept SQL queries and returns the response in JSON format.

Brokers need three key things to start.

• Cluster name

• Zookeeper address

• Broker instance name

At the start, a broker registers as a Helix Participant and awaits notifications from other Helix agents. These notifications will be handled for table creation, a new segment being loaded, or a server starting up/or going down, in addition to any configuration changes.

Service Discovery/Routing Table

Irrespective of the kind of notification, the key responsibility of a broker is to maintain the query routing table. The query routing table is simply a mapping between segments and the servers that a segment resides on. Typically, a segment resides on more than one server. The broker computes multiple routing tables depending on the configured routing strategy for a table. The default strategy is to balance the query load across all available servers.

//This is an example ZNode config for EXTERNAL VIEW in Helix
{
  "id" : "baseballStats_OFFLINE",
  "simpleFields" : {
    ...
  },
  "mapFields" : {
    "baseballStats_OFFLINE_0" : {
      "Server_10.1.10.82_7000" : "ONLINE"
    }
  },
  ...
}

Query processing

For every query, a cluster's broker performs the following:

• Fetches the routes that are computed for a query based on the routing strategy defined in a table's configuration.

• Computes the list of segments to query from on each server.

• Scatter-Gather: sends the requests to each server and gathers the responses.

• Merge: merges the query results returned from each server.

• Sends the query result to the client.

// Query: select count(*) from baseballStats limit 10

// RESPONSE
// ========
{
    "resultTable": {
        "dataSchema": {
            "columnDataTypes": ["LONG"],
            "columnNames": ["count(*)"]
        },
        "rows": [
            [97889]
        ]
    },
    "exceptions": [],
    "numServersQueried": 1,
    "numServersResponded": 1,
    "numSegmentsQueried": 1,
    "numSegmentsProcessed": 1,
    "numSegmentsMatched": 1,
    "numConsumingSegmentsQueried": 0,
    "numDocsScanned": 97889,
    "numEntriesScannedInFilter": 0,
    "numEntriesScannedPostFilter": 0,
    "numGroupsLimitReached": false,
    "totalDocs": 97889,
    "timeUsedMs": 5,
    "segmentStatistics": [],
    "traceInfo": {},
    "minConsumingFreshnessTimeMs": 0
}

Fault tolerance

Broker instances scale horizontally without an upper bound. In a majority of cases, only three brokers are required. If most query results that are returned to a client are <1MB in size per query, one can run a broker and servers inside the same instance container. This lowers the overall footprint of a cluster deployment for use cases that do not need to guarantee a strict SLA on query performance in production.

Server

Servers host segments and do most of the heavy lifting during query processing. Though the architecture shows that there are two kinds of servers, real-time and offline, a server does not really know if it's going to be a real-time server or an offline server. The responsibility of a server depends on the table assignment strategy.

Offline servers

Offline servers typically host segments that are immutable. In this case, segments are created outside of a cluster and uploaded via a shell-based curl request. Based on the replication factor and the segment assignment strategy, the controller picks one or more servers to host the segment. Servers are notified via Helix about the new segments. Servers fetch the segments from deep store and loads them before being ready to serve query requests. At this point, the cluster's broker detects that new segments are available and starts including them in query responses.

Real-time servers

Real-time servers are different from the offline servers. Real-time server nodes ingest data from streaming sources, such as Kafka, and generate the indexed segments in-memory (flushing segments to disk periodically). In memory segments are also known as consuming segments. These consuming segments get flushed periodically based on completion threshold (based on number of rows, time or segment size). At this point, they are known as completed segments. Completed segments are similar to the offline server's segments. Queries go over the in-flight (consuming) segments and the completed segments.

Minion

Minion is an optional component and is not required to get started with Pinot. Minion is used for purging data from a Pinot cluster (for reasons such as GDPR compliance in the UK).

Data ingestion overview

Within Pinot, a logical table is modeled as one of two types of physical tables: offline or real-time. The reason for having two types of tables is because each one follows a different state model.

A real-time and offline table provide different configuration options for indexing and, in the case of real-time, the connector properties for the stream data source (i.e. Kafka). Table types also allow users to use different containers for real-time and offline server nodes. For instance, offline servers might use virtual machines with larger storage capacity where real-time servers might need higher system memory and/or more CPU cores.

The two types of tables also scale differently.

• Real-time tables have a smaller retention period and scales query performance based on the ingestion rate.

• Offline tables have larger retention and scales performance based on the size of stored data.

There are a few things to keep in mind when configuring the different types of tables for your workloads. When ingesting data from the same source, you can have two tables that ingest the same data that are configured differently for real-time and offline queries. Even though the two tables have the same data, performance will scale differently for queries based on your requirements. In this scenario, real-time and offline tables must share the same schema.

Batch data flow

In batch mode, data is ingested into Pinot via an ingestion job. An ingestion job transforms a raw data source (such as a CSV file) into segments. Once segments are generated for the imported data, an ingestion job stores them into the cluster's segment store (a.k.a deep store) and notifies the controller. The notification is processed and the result is that the Helix agent on the controller updates the ideal state configuration in Zookeeper. Helix will then notify the offline server that there are new segments available. In response to the notification from the controller, the offline server downloads the newly created segments directly from the cluster's segment store. The cluster's broker, which watches for state changes in Helix, detects the new segments and adds them to the list of segments to query (segment-to-server routing table).

Real-time data flow

At table creation, a controller creates a new entry in Zookeeper for the consuming segment. Helix notices the new segment and notifies the real-time server, which start consuming data from the streaming source. The broker, which watches for changes, detects the new segments and adds them to the list of segments to query (segment-to-server routing table).

Whenever the segment is complete (i.e. full), the real-time server notifies the Controller, which checks with all replicas and picks a winner to commit the segment to. The winner commits the segment and uploads it to the cluster's segment store, updating the state of the segment from "consuming" to "online". The controller then prepares a new segment in a "consuming" state.

Query overview

Queries are received by brokers—which checks the request against the segment-to-server routing table—scattering the request between real-time and offline servers.

The two tables then process the request by filtering and aggregating the queried data, which is then returned back to the broker. Finally, the broker gathers together all of the pieces of the query response and responds back to the client with the result.

Cluster is a set a nodes comprising of servers, brokers, controllers and minions.

Pinot leverages Apache Helix for cluster management. Helix is a cluster management framework to manage replicated, partitioned resources in a distributed system. Helix uses Zookeeper to store cluster state and metadata.

Cluster components

Briefly, Helix divides nodes into three logical components based on their responsibilities

Participant

The nodes that host distributed, partitioned resources

Spectator

The nodes that observe the current state of each Participant and use that information to access the resources. Spectators are notified of state changes in the cluster (state of a participant, or that of a partition in a participant).

Controller

The node that observes and controls the Participant nodes. It is responsible for coordinating all transitions in the cluster and ensuring that state constraints are satisfied while maintaining cluster stability.

Pinot Servers are modeled as Participants, more details about server nodes can be found in Server. Pinot Brokers are modeled as Spectators, more details about broker nodes can be found in Broker. Pinot Controllers are modeled as Controllers, more details about controller nodes can be found in Controller.

Logical view

Another way to visualize the cluster is a logical view, wherein a cluster contains tenants, tenants contain tables, and tables contain segments.

Setup a Pinot Cluster

Typically, there is only cluster per environment/data center. There is no needed to create multiple Pinot clusters since Pinot supports the concept of tenants. At LinkedIn, the largest Pinot cluster consists of 1000+ nodes.

To setup a Pinot cluster, we need to first start Zookeeper.

docker network create -d bridge pinot-demo

docker run \
    --network=pinot-demo \
    --name pinot-zookeeper \
    --restart always \
    -p 2181:2181 \
    -d zookeeper:3.5.6

docker run \
    --network pinot-demo --name=zkui \
    -p 9090:9090 \
    -e ZK_SERVER=pinot-zookeeper:2181 \
    -d qnib/plain-zkui:latest

bin/pinot-admin.sh StartZookeeper -zkPort 2181

Once we've started Zookeeper, we can start other components to join this cluster. If you're using docker, pull the latest apachepinot/pinot image.

export PINOT_VERSION=0.3.0-SNAPSHOT
export PINOT_IMAGE=apachepinot/pinot:${PINOT_VERSION}
docker pull ${PINOT_IMAGE}

To start other components to join the cluster

1. Start Controller

2. Start Broker

3. Start Server

Explore your cluster via Pinot Data Explorer

Servers host the data segments and serve queries off the data they host. There's two types of servers

Offline Offline servers are responsible for downloading segments from the segment store, to host and serve queries off. When a new segment is uploaded to the controller, the controller decides the servers (as many as replication) that will host the new segment and notifies them to download the segment from the segment store. On receiving this notification, the servers download the segment file and load the segment onto the server, to server queries off them.

Realtime Real time servers directly ingest from a real time stream (such as Kafka, EventHubs). Periodically, they make segments of the in-memory ingested data, based on certain thresholds. This segment is then persisted onto the segment store.

Pinot Servers are modeled as Helix Participants, hosting Pinot tables (referred to as resources in helix terminology). Segments of a table are modeled as Helix partitions (of a resource). Thus, a Pinot server hosts one or more helix partitions of one or more helix resources (i.e. one or more segments of one or more tables).

Starting a Server

Make sure you've setup Zookeeper. If you're using docker, make sure to pull the pinot docker image. To start a server

Usage: StartServer
    -serverHost               <String>                      : Host name for controller. (required=false)
    -serverPort               <int>                         : Port number to start the server at. (required=false)
    -serverAdminPort          <int>                         : Port number to serve the server admin API at. (required=false)
    -dataDir                  <string>                      : Path to directory containing data. (required=false)
    -segmentDir               <string>                      : Path to directory containing segments. (required=false)
    -zkAddress                <http>                        : Http address of Zookeeper. (required=false)
    -clusterName              <String>                      : Pinot cluster name. (required=false)
    -configFileName           <Config File Name>            : Broker Starter Config file. (required=false)
    -help                                                   : Print this message. (required=false)

>

docker run \
    --network=pinot-demo \
    --name pinot-server \
    -d ${PINOT_IMAGE} StartServer \
    -zkAddress pinot-zookeeper:2181

bin/pinot-admin.sh StartServer \
    -zkAddress localhost:2181

USAGE

Usage: StartServer
    -serverHost               <String>                      : Host name for controller. (required=false)
    -serverPort               <int>                         : Port number to start the server at. (required=false)
    -serverAdminPort          <int>                         : Port number to serve the server admin API at. (required=false)
    -dataDir                  <string>                      : Path to directory containing data. (required=false)
    -segmentDir               <string>                      : Path to directory containing segments. (required=false)
    -zkAddress                <http>                        : Http address of Zookeeper. (required=false)
    -clusterName              <String>                      : Pinot cluster name. (required=false)
    -configFileName           <Config File Name>            : Server Starter Config file. (required=false)
    -help                                                   : Print this message. (required=false)

Brokers are the components that handle Pinot queries. They accept queries from clients and forward them to the right servers. They collect results back from the servers and consolidate them into a single response, to sent it back to the client.

Pinot Brokers are modeled as Spectators. They need to know the location of each segment of a table (and each replica of the segments) and route requests to the appropriate server that hosts the segments of the table being queried. The broker ensures that all the rows of the table are queried exactly once so as to return correct, consistent results for a query. The brokers may optimize to prune some of the segments as long as accuracy is not sacrificed. Helix provides the framework by which spectators can learn the location in which each partition of a resource (i.e. participant) resides. The brokers use this mechanism to learn the servers that host specific segments of a table.

In case of hybrid tables, the brokers ensure that the overlap between realtime and offline segment data is queried exactly once, by performing offline and realtime federation. Let's take this example, we have realtime data for 5 days - March 23 to March 27, and offline data has been pushed until Mar 25, which is 2 days behind realtime. The brokers maintain this time boundary.

Suppose, we get a query to this table : select sum(metric) from table. The broker will split the query into 2 queries based on this time boundary - one for offline and one for realtime. This query becomes - select sum(metric) from table_REALTIME where date >= Mar 25 and select sum(metric) from table_OFFLINE where date < Mar 25 The broker then merges results from both these queries before returning back to the client.

Starting a Broker

Make sure you've setup Zookeeper. If you're using docker, make sure to pull the pinot docker image. To start a broker

docker run \
    --network=pinot-demo \
    --name pinot-broker \
    -d ${PINOT_IMAGE} StartBroker \
    -zkAddress pinot-zookeeper:2181

bin/pinot-admin.sh StartBroker \
  -zkAddress localhost:2181 \
  -clusterName PinotCluster \
  -brokerPort 7000

A tenant is a logical component, defined as a group of server/broker nodes with the same Helix tag.

In order to support multi-tenancy, Pinot has first class support for tenants. Every table is associated with a server tenant and a broker tenant. This controls the nodes that will be used by this table as servers and brokers. This allows all tables belonging to a particular use case to be grouped under a single tenant name.

The concept of tenants is very important when the multiple use cases are using Pinot and there is a need to provide quotas or some sort of isolation across tenants. For example, consider we have two tables Table A and Table B in the same Pinot cluster.

We can configure Table A with server tenant Tenant A and Table B with server tenant Tenant B. We can tag some of the server nodes for Tenant A and some for Tenant B. This will ensure that segments of Table A only reside on servers tagged with Tenant A, and segment of Table B only reside on servers tagged with Tenant B. The same isolation can be achieved at the broker level, by configuring broker tenants to the tables.

No need to create separate clusters for every table or use case!

Tenant Config

This tenant is defined in the tenants section of the table config.

This section contains 2 main fields broker and server which decide the tenants used for the broker and server components of this table.

"tenants": {
  "broker": "brokerTenantName",
  "server": "serverTenantName"
}

In the above example,

• The table will be served by brokers that have been tagged as brokerTenantName_BROKER in Helix.

• If this were an offline table, the offline segments for the table will be hosted in pinot servers tagged in helix as serverTenantName_OFFLINE

• If this were a realtime table, the realtime segments (both consuming as well as completed ones) will be hosted in pinot servers tagged in helix as serverTenantName_REALTIME.

Creating a tenant

Broker tenant

Here's a sample broker tenant config. This will create a broker tenant sampleBrokerTenant by tagging 3 untagged broker nodes as sampleBrokerTenant_BROKER.

{
     "tenantRole" : "BROKER",
     "tenantName" : "sampleBrokerTenant",
     "numberOfInstances" : 3
}

To create this tenant use the following command. The creation will fail if number of untagged broker nodes is less than numberOfInstances.

bin/pinot-admin.sh AddTenant \
    -name sampleBrokerTenant 
    -role BROKER 
    -instanceCount 3 -exec

curl -i -X POST -H 'Content-Type: application/json' -d @sample-broker-tenant.json localhost:9000/tenants

Check out the table config in the Rest API to make sure it was successfully uploaded.

Server tenant

Here's a sample server tenant config. This will create a server tenant sampleServerTenant by tagging 1 untagged server node as sampleServerTenant_OFFLINE and 1 untagged server node as sampleServerTenant_REALTIME.

{
     "tenantRole" : "SERVER",
     "tenantName" : "sampleServerTenant",
     "offlineInstances" : 1,
     "realtimeInstances" : 1
}

To create this tenant use the following command. The creation will fail if number of untagged server nodes is less than offlineInstances + realtimeInstances.

bin/pinot-admin.sh AddTenant \
    -name sampleServerTenant \
    -role SERVER \
    -offlineInstanceCount 1 \
    -realtimeInstanceCount 1 -exec

curl -i -X POST -H 'Content-Type: application/json' -d @sample-server-tenant.json localhost:9000/tenants

Check out the table config in the Rest API to make sure it was successfully uploaded.

Schema is used to define the names, data types and other information for the columns of a Pinot table.

Types of columns

Columns in a Pinot table can be broadly categorized into three categories

Schema format

A Pinot schema is written in JSON format. Here's an example which shows all the fields of a schema

The Pinot schema is composed of

Below is a detailed description of each type of field spec.

dimensionFieldSpecs

A dimensionFieldSpec is defined for each dimension column. Here's a list of the fields in the dimensionFieldSpec

Internal default null values for dimension

metricFieldSpecs

A metricFieldSpec is defined for each metric column. Here's a list of fields in the metricFieldSpec

Internal default null values for metric

dateTimeFieldSpec

A dateTimeFieldSpec is used to define time columns of the table. Here's a list of the fields in a dateTimeFieldSpec

timeFieldSpec

This has been deprecated. Older schemas containing timeFieldSpec will be supported. But for new schemas, use DateTimeFieldSpec instead.

A timeFieldSpec is defined for the time column. A timeFieldSpec is composed of an incomingGranularitySpec and an outgoingGranularitySpec. IncomingGranularitySpec in combination with outgoingGranularitySpec can be used to transform the time column from incoming format to the outgoing format. If both of them are specified, the segment creation process will convert the time column from the incoming format to the outgoing format. If no time column transformation is required, you can specify just the incomingGranularitySpec.

The incoming and outgoing granularitySpec are defined as:

Advanced fields

Apart from these, there's some advanced fields. These are common to all field specs.

Ingestion Transform Functions

Transform functions can be defined on columns in the schema. For example:

Currently, we have support for 2 kinds of functions

1. Groovy functions

2. Inbuilt functions

Groovy functions

Groovy functions can be defined using the syntax:

Here's some examples of commonly needed functions. Any valid Groovy expression can be used.

String concatenation

Concat firstName and lasName to get fullName

Find element in an array

Find max value in array bids

Time transformation

Convert timestamp from MILLISECONDS to HOURS

Column name change

Simply change name of the column from user_id to userId

Ternary operation

If eventType is IMPRESSION set impression to 1. Similar for CLICK.

AVRO Map

Store an AVRO Map in Pinot as two multi-value columns. Sort the keys, to maintain the mapping. 1) The keys of the map as map_keys 2) The values of the map as map_values

Inbuilt Pinot functions

We have several inbuilt functions that can be used directly in as ingestion transform functions

DateTime functions

These are functions which enable commonly needed time transformations.

toEpochXXX

Converts from epoch milliseconds to a higher granularity.

toEpochXXXRounded

Converts from epoch milliseconds to another granularity, rounding to the nearest rounding bucket. For example, 1588469352000 (2020-05-01 42:29:12) is 26474489 minutesSinceEpoch. `toEpochMinutesRounded(1588469352000) = 26474480 (2020-05-01 42:20:00)

fromEpochXXX

Converts from an epoch granularity to milliseconds.

Simple date format

Converts simple date format strings to milliseconds and vice-a-versa, as per the provided pattern string.

Json functions

Creating a Schema

Running Pinot

We want your experience getting started with Pinot to be both low effort and high reward. Here you'll find a collection of quick start guides that contain starter distributions of the Pinot platform.

Bootstrapping a cluster

Deploy to a public cloud

How to setup a Pinot cluster

This video will show you a step-by-step walk through for launching the individual components of Pinot and scaling them to multiple instances. This is an excellent resource for developers and operators that want to understand setting up each component and debugging a cluster.

We also have a step-by-step guide for manually setting up a Pinot cluster using Docker or shell scripts.

Data import examples

In this guide you'll learn how to download and install Apache Pinot as a standalone instance.

Download Apache Pinot

First, let's download the Pinot distribution for this tutorial. You can either build the distribution from source or download a packaged release.

Build from source or download the distribution

Setting up a Pinot cluster

We'll be using a quick-start script, which does the following:

1. Sets up the Pinot cluster QuickStartCluster

2. Creates a sample table and loads sample data

There's 3 kinds of quick start

Batch

Batch quick start creates the pinot cluster, creates an offline table baseballStats and pushes sample offline data to the table.

Streaming

Streaming quick start sets up a Kafka cluster and pushes sample data to a Kafka topic. Then, it creates the Pinot cluster and creates a realtime table meetupRSVP which ingests data from the Kafka topic.

Hybrid

Hybrid quick start sets up a Kafka cluster and pushes sample data to a Kafka topic. Then, it creates the Pinot cluster and creates a hybrid table airlineStats . The realtime table ingests data from the Kafka topic. Lastly, sample data is pushed into the offline table.

Ingestion

How do I flatten my JSON Kafka stream?

We have toJsonStr(key) function which can store a top level json field as a STRING in Pinot.

Then you can use jsonExtractScalar(JSON_STRING_FIELD, JSON_PATH, OUTPUT_FORMAT) function during query time to fetch the desired field from the json string. For example

Indexing

How to set inverted indexes?

How to apply inverted index to existing setup?

How to apply star tree index?

Querying

What are all the fields in the Pinot query's JSON response?

SQL Query fails with "Encountered 'timestamp' was expecting one of..."

"timestamp" is a reserved keyword in SQL. Escape timestamp with double quotes.

Other commonly encountered reserved keywords are date, time, table.

Filtering on STRING column WHERE column = "foo" does not work?

For filtering on STRING columns, use single quotes

ORDER BY using an alias doesn't work?

The fields in the ORDER BY clause must be one of the group by clauses or aggregations, BEFORE applying the alias. Therefore, this will not work

Instead, this will work

Operations

Can I change a column name in my table, without losing data?

How to change number of replicas of a table?

How to run a rebalance on a table?

A rebalance is run to reassign all the segments of a table to the available servers. This is typically done when capacity changes are done i.e. adding more servers or removing servers from a table.

Offline

Realtime

You can check the status of the rebalance by

1. Checking the controller logs

2. Running rebalance again after a while, you should receive status "status": "NO_OP"

3. Checking the External View of the table, to see the changes in capacity/replicas have taken effect.

Tuning and Optimizations

Do replica groups work for real-time?

Yes, replica groups work for realtime. There's 2 parts to enabling replica groups:

1. Replica groups segment assignment

2. Replica group query routing

Replica group segment assignment

Replica group segment assignment is achieved in realtime, if number of servers is a multiple of number of replicas. The partitions get uniformly sprayed across the servers, creating replica groups.

For example, consider we have 6 partitions, 2 replicas, and 4 servers.

Replica group query routing

Create an isolated bridge network in docker

We'll be using our docker image apachepinot/pinot:latest to run this quick start, which does the following:

• Sets up the Pinot cluster

• Creates a sample table and loads sample data

There are 3 types of quick start examples.

• Batch example

• Streaming example

• Hybrid example

Batch example

In this example we demonstrate how to do batch processing with Pinot.

• Starts Pinot deployment by starting

  • Apache Zookeeper

  • Pinot Controller

  • Pinot Broker

  • Pinot Server

• Creates a demo table

  • baseballStats

• Launches a standalone data ingestion job

  • Builds one Pinot segment for a given CSV data file for table baseballStats

  • Pushes the built segment to the Pinot controller

• Issues sample queries to Pinot

Once the Docker container is running, you can view the logs by running the following command.

That's it! We've spun up a Pinot cluster.

Your cluster is ready once you see the cluster setup completion messages and sample queries, as demonstrated below.

Streaming example

In this example we demonstrate how to do stream processing with Pinot.

• Starts Pinot deployment by starting

  • Apache Kafka

  • Apache Zookeeper

  • Pinot Controller

  • Pinot Broker

  • Pinot Server

• Creates a demo table

  • meetupRsvp

• Launches a meetup **stream

• Publishes data to a Kafka topic meetupRSVPEvents to be subscribed to by Pinot

• Issues sample queries to Pinot

Hybrid example

In this example we demonstrate how to do hybrid stream and batch processing with Pinot.

1. Starts Pinot deployment by starting

   • Apache Kafka

   • Apache Zookeeper

   • Pinot Controller

   • Pinot Broker

   • Pinot Server

2. Creates a demo table

   • airlineStats

3. Launches a standalone data ingestion job

   • Builds Pinot segments under a given directory of Avro files for table airlineStats

   • Pushes built segments to Pinot controller

4. Launches a **stream of flights stats

5. Publishes data to a Kafka topic airlineStatsEvents to be subscribed to by Pinot

6. Issues sample queries to Pinot

Pinot has the concept of , which is a logical abstraction to refer to a collection of related data. Pinot has a distributed architecture and scales horizontally. Pinot expects the size of a table to grow infinitely over time. In order to achieve this, the entire data needs to be distributed across multiple nodes. Pinot achieve this by breaking the data into smaller chunks known as segment (this is similar to shards/partitions in relational databases). Segments can also be seen as time based partitions.

Thus, a segment is a horizontal shard representing a chunk of table data with some number of rows. The segment stores data for all columns of the table. Each segment packs the data in a columnar fashion, along with the dictionaries and indices for the columns. The segment is laid out in a columnar format so that it can be directly mapped into memory for serving queries.

Columns may be single or multi-valued. Column types may be STRING, INT, LONG, FLOAT, DOUBLE or BYTES. Columns may be declared to be metric or dimension (or specifically as a time dimension) in the schema. Columns can have default null value. For example, the default null value of a integer column can be 0. Note: The default value of byte column has to be hex-encoded before adding to the schema.

Pinot uses dictionary encoding to store values as a dictionary ID. Columns may be configured to be “no-dictionary” column in which case raw values are stored. Dictionary IDs are encoded using minimum number of bits for efficient storage (e.g. a column with cardinality of 3 will use only 2 bits for each dictionary ID).

There is a forward index built for each column and compressed appropriately for efficient memory use. In addition, optional inverted indices can be configured for any set of columns. Inverted indices, while take up more storage, offer better query performance. Specialized indexes like Star-Tree index is also supported. Check out for more details.

Creating a segment

Once the table is configured, we can load some data. Loading data involves generating pinot segments from raw data and pushing them to the pinot cluster. Data can be loaded in batch mode or streaming mode. See page for details.

Load Data in Batch

Prerequisites

Below are instructions to generate and push segments to Pinot via standalone scripts. For a production setup, you should use frameworks such as Hadoop or Spark. See this for more details on setting up Data Ingestion Jobs.

Job Spec YAML

To generate a segment, we need to first create a job spec yaml file. JobSpec yaml file has all the information regarding data format, input data location and pinot cluster coordinates. Note that this assumes that the controller is RUNNING to fetch the table config and schema. If not, you will have to configure the spec to point at their location.

where,

executionFrameworkSpec

pinotFSSpecs

recordReaderSpec

tableSpec

segmentNameGeneratorSpec

pinotClusterSpecs

pushJobSpec

Create and push segment

To create and push the segment in one go, use

Alternately, you can separately create and then push, by changing the jobType to SegmentCreation or SegmenTarPush.

Templating Ingestion Job Spec

Ingestion job spec supports templating with Groovy Syntax.

This would be convenient for users to generate one ingestion job template file and schedule it in a daily basis with extra parameters updated daily.

E.g. users can set inputDirURI with parameters to indicate date, so that ingestion job only process the data for a particular date.

Below is an example to specify the date templating for input and output path.

Then specify the value of ${year}, ${month}, ${day} when kicking off the ingestion job with arguments: -values $param=value1 $param2=value2...

This ingestion job only generates segment for date 2014-01-03

Load Data in Streaming

Prerequisites

Below is an example of how to publish sample data to your stream. As soon as data is available to the realtime stream, it starts getting consumed by the realtime servers

Kafka