The evolving HBase ecosystem

A few months back we made an exhaustive list of open source Apache Hadoop ecosystem projects. Continuing with the interest in ecosystem awareness, let’s take a deeper dive in to commercial and open source products evolving around HBase. The list below does not look at use cases; rather it focuses on products which can be used by developers, data scientists and analysts. Each product section lists in brief how it is leveraging HBase in its architecture.

Lily:

Initially built as a big data based content repository, Lily (NGDATA) today is a powerful, unified, integrated solution that uniquely combines an interactive Big Data management platform and consumer intelligence applications. The current Lily solution architecture consists of three key layers: Big Data repository based on Apache Hadoop and HBase; consumer database; customer intelligence applications. “Lily HBase Indexer provides the ability to quickly and easily search for any content stored in HBase. It allows you to quickly and easily index HBase rows into Solr, without writing a line of code. Lily HBase Indexer is an open source project available stand-alone, or as part of Cloudera Search.”

Cloudera Search:

Cloudera Search leverages the Lily HBase Indexer Service for “a flexible, scalable, fault tolerant, transactional, Near Real Time (NRT) oriented system for processing a continuous stream of HBase cell updates into live search indexes … The Lily HBase Indexer uses SolrCloud to index data stored in HBase. As HBase applies inserts, updates, and deletes to HBase table cells, the indexer keeps Solr consistent with the HBase table contents, using standard HBase replication features. The indexer supports flexible custom application-specific rules to extract, transform, and load HBase data into Solr. Solr search results can contain columnFamily:qualifier links back to the data stored in HBase. This way, applications can use the Search result set to directly access matching raw HBase cells.”

Cloudera Impala:

Impala provides analytical tool to query HBase tables. Impala by default uses data files stored on HDFS. However, it can be used to query HBase for OLTP-style workloads, with lookups of individual rows or ranges of values. Impala uses the HBase client API via Java Native Interface (JNI) to query data stored in HBase. It is interesting to note that Impala and Hive share the same metastore database which means once can create HBase tables on the Impala side using the Hive shell.

Apache Drill:

Drill supports querying against many different schema-less data sources including HBase, Cassandra and MongoDB. Further it provides a HBase storage engine as an implementation of default storage engine. Drill uses a parser API to transform the query into logical operators. Futher, with the help of optimizer and execution agents called foreman and HBase scanner it reads entries from HTable.

Continuuity Reactor:

Co-founded by Jonathan Gray, Continuuity as per Wired magazine is gifting the social network’s data platform to the world. It aims to abstract the complexity of building and deploying Hadoop applications. Built on top of MapReduce engine and HBase, it provides methods for processing data in real time as well. The Continuuity DataFabric is leverages HBase to stores data which is accessed by multiple apps on a Reactor or via REST. The HBase data is manipulated with Datasets which consists of data and provides methods to manipulate it via an API.

Hannibal:

Hannibal is tool to help monitor and maintain HBase-Clusters that are configured for manual splitting. As per Nils Kübler, the creator of the Hannibal project, “It widens the monitoring capabilities of HBase by providing different views with interactive graphs of the cluster. Hannibal is also a Web-based tool that fits smoothly into your existing Hadoop/HBase ecosystem.”

Kiji:

Kiji from WibiData is probably one of most vocal open source projects around HBase extension. Built on Apache HBase and Apache Hadoop, the open source Kiji Project provides a framework for building big data applications. It provides components that store, analyze, and serve recommendations and other machine learning / analytic results in real time.

It “provides a model and a set of libraries that allow developers to get up and running quickly.  Intuitive Java APIs and Kiji’s rich data model allow developers to build business logic and machine learning algorithms without having to worry about bytes, serialization, schema evolutions and lower-level aspects of the system. The Kiji framework is modularized into separate components to support a wide range of usage and encourage clean separation of functionality.” Kiji is offered in a packaged BentoBox which contains following utilities for leveraging HBase in various modules like KijiScoring:

* bento-cluster - a single-process Hadoop/HBase/ZK cluster embedded in the BentoBox.

* fake-hbase - An in-memory implementation of HBase without persistence, used for unit tests.

*  hbase-maven-plugin - A Maven plugin that hosts a MiniHBase/MiniZK/MiniMR/MiniHDFS cluster used for integration tests

Phoenix:

Phoenix is an open source SQL skin for HBase promoted by Salesforce. It uses standard JDBC APIs instead of the regular HBase client APIs to create tables, insert data, and query HBase data. Per James Taylor, “Phoenix compiles your SQL query into a series of HBase scans, and orchestrates the running of those scans to produce regular JDBC result sets. On top of that, you can add secondary indexes to your tables to transform what would normally be a full table scan into a series of point gets (and we all know how good HBase performs with those).”

OpenTSDB:

OpenTSDB is a distributed, scalable Time Series Database (TSDB) written on top of HBase which helps to store, index and serve metrics collected from computer systems at a large scale.  Like Ganglia, OpenTSDB can be used to monitor various systems including HBase.

Splice Machine:

Splice Machine integrates Apache Derby and Apache HBase technology stacks by replacing the storage engine in Derby with HBase. It retains the Apache Derby parser but has redesigned the planner, optimizer, and executor to leverage the distributed HBase computation engine.

“HBase co-processors are used to embed Splice Machine in each distributed HBase region (i.e., data shard). This enables Splice Machine to achieve massive parallelization by pushing the computation down to each distributed data shard.”

Spire:

Spire from Drawn to Scale has an indexing, schema, and query engine on top of the HBase database. Further with MapR’s HBase-compatible distributed filesystem embedded in Spire, it has attempted to build a real time HBase datastore.

“By running Spire and therefore HBase on top of MapR we take advantage of their optimized C++ file server that is highly asynchronous.  Combined with an multiplexed socket reuse model, Spire dispatches many random read requests at once without having to worry about socket-per-read or thread-per-read and running out of resources.  This allows the MapR file server to handle thousands or even tens of thousands open files and concurrent disk operations.”

Scaled Risk:

Built on top of HBase and HDFS, Scaled Risk guarantees integrity of transactions. It claims to capture all information from enterprise systems and make it persist forever. It also claims to make data available as soon as published, in real-time.

IBM BigSQL:

BigSQL allows users to run point and big ad-hoc queries on HBase tables with SQL. With JDBC and ODBC compliance, data can be imported or tables manipulated by various applications running on different OS. Further, analytical tools like Cognos can be used to import, analyze and visualize data.