ENTRIES TAGGED "Hadoop query"
The inaugural Spark Summit will feature a wide variety of real-world applications
When an interesting piece of big data technology gets introduced, early1 adopters tend to focus on technical features and capabilities. Applications get built as companies develop confidence that it’s reliable and that it really scales to large data volumes. That seems to be where Spark is today. With over 90 contributors from 25 companies, it has one of the largest developer communities among big data projects (second only to Hadoop MapReduce).
I recently became an advisor to Databricks (a startup commercializing Spark) and a member of the program committee for the inaugural Spark Summit. As I pored over submissions to Spark’s first community gathering, I learned how companies have come to rely on Spark, Shark, and other components of the Berkeley Data Analytics Stack (BDAS). Spark is at that stage where companies are deploying it, and the upcoming Spark Summit in San Francisco will showcase many real-world applications. These applications cut across many domains including advertising, marketing, finance, and academic/scientific research, but can generally be grouped into the following categories:
Data processing workflows: ETL and Data Wrangling
Many companies rely on a wide variety of data sources for their analytic products. That means cleaning, transforming, and fusing (unstructured) external data with internal data sources. Many companies – particularly startups – use Spark for these types of data processing workflows. There are even companies that have created simple user interfaces that open up batch data processing tasks to non-programmers.
Tools for unlocking big data continue to get simpler
Here are a few observations based on conversations I had during the just concluded Strata NYC conference.
Interactive query analysis on Hadoop remains a hot area
A recent O’Reilly survey confirmed SQL is an important skill for data scientists. A year after the launch of Impala, quite a few attendees I spoke with remained interested in the progress of SQL-on-Hadoop solutions. A trio from Hortonworks gave an update on recent improvements and changes to Hive1. A sign that Impala is gaining traction, Greg Rahn’s talk on Practical Performance Tuning for Impala was one of the best attended sessions in the conference. Ditto for a sponsored session on Kognitio’s latest features.
Existing SQL-on-Hadoop solutions require that users define a schema – an additional step given that a lot of data is increasingly in key-value or JSON format. In his talk Hadapt co-founder Daniel Abadi highlighted a solution2 that lets users query complex data types (Hadapt reserializes complex data types to speed up joins). I expect other SQL-on-Hadoop solutions to also offer query support for complex data types in the near future.
Empowering business users
With its launch at the conference, ClearStory joins Platfora and Datameer in the business analytics space. Each company builds tools that lets business users wade through large amounts of data, while emphasizing different areas. Platfora is for interactive visual analysis of massive data sets, while Datameer connects to many data sources (not just Hadoop), has started offering analytics, and can run on a laptop or cluster. Built primarily on the Berkeley stack (BDAS), ClearStory’s interesting platform encourages collaboration and simplifies data harmonization (fusing disparate data sources is a common bottleneck for business users). For organizations willing to tag and describe their data sets, Microsoft unveiled a tool that lets users query data using natural language (UK startup NeutrinoBI uses a similar “search interface”).
Data stores are rolling out easy-to-use analysis tools
Originated by the NSA, Apache Accumulo is a BigTable inspired data store known for being highly scalable and for its interesting security model. Federal agencies and Defense contractors have deployed Accumulo on clusters of a thousand or more servers. It also uses “cell-level” security to control access to values stored in individual cells1.
What Accumulo was lacking were easy-to-use, standard analytic engines that allow users to interact with data. The release of Sqrrl Enterprise this past week fills that gap. Sqrrl Enterprise provides an initial set of analytic engines for the Accumulo ecosystem2. It includes support for interactive SQL, fulltext search, and queries over graph data. Each of these engines takes into account security labels placed on data: since every data object ingested into Sqrrl has a security label, (query & analytic) results incorporate those access levels. Analysts interact with data as they normally would. For example Sqrrl’s indexing technology accounts for security labels, and search queries are written in standard Lucene syntax. Reminiscent of the Phoenix project for HBase3, SQL queries4 in Sqrrl are converted into optimized Accumulo iterators.
New open source tools for interactive SQL analysis, model development and deployment
When Hadoop users need to develop apps that are “latency sensitive”, many of them turn to HBase1. Its tight integration with Hadoop makes it a popular data store for real-time applications. When I attended the first HBase conference last year, I was pleasantly surprised by the diversity of companies and applications that rely on HBase. This year’s conference was even bigger and I ran into attendees from a wide range of companies. Another set of interesting real-world case studies were showcased, along with sessions highlighting work of the HBase team aimed at improving usability, reliability, and availability (bringing down mean time to recovery has been a recent area of focus).
HBase has had a reputation of being a bit difficult to use – its core users have been data engineers, not data scientists. The good news is that as HBase gets adopted by more companies, tools are being developed to open it up to more users. Let me highlight some tools that will appeal to data scientists.
Analytic engines on top of Hadoop simplify the creation of interesting, low-cost, scalable applications
Hadoop’s low-cost, scale-out architecture has made it a new platform for data storage. With a storage system in place, the Hadoop community is slowly building a collection of open source, analytic engines. Beginning with batch processing (MapReduce, Pig, Hive), Cloudera has added interactive SQL (Impala), analytics (Cloudera ML + a partnership with SAS), and as of early this week, real-time search. The economics that led to Hadoop dominating batch processing is permeating other types of analytics.
Another collection of open source, Hadoop-compatible analytic engines, the Berkeley Data Analytics Stack (BDAS), is being built just across the San Francisco Bay. Starting with a batch-processing framework that’s faster than MapReduce (Spark), it now includes interactive SQL (Shark), and real-time analytics (Spark Streaming). Sometime this summer, frameworks for machine-learning (MLbase) and graph analytics (GraphX) will be released. A cluster manager (Mesos) and an in-memory file system (Tachyon) allow users of other analytic frameworks to leverage the BDAS platform. (The Python data community is looking at Tachyon closely.)
A new, open source benchmark can be used to track performance improvements over time
As organizations continue to accumulate data, there has been renewed interest in interactive query engines that scale to terabytes (even petabytes) of data. Traditional MPP databases remain in the mix, but other options are attracting interest. For example, companies willing to upload data into the cloud are beginning to explore Amazon Redshift1, Google BigQuery, and Qubole.
A variety of analytic engines2 built for Hadoop are allowing companies to bring its low-cost, scale-out architecture to a wider audience. In particular, companies are rediscovering that SQL makes data accessible to lots of users, and many prefer3 not having to move data to a separate (MPP) cluster. There are many new tools that seek to provide an interactive SQL interface to Hadoop, including Cloudera’s Impala, Shark, Hadapt, CitusDB, Pivotal-HD, PolyBase4, and SQL-H.
An open source benchmark from UC Berkeley’s Amplab
A benchmark for tracking the progress5 of scalable query engines has just been released. It’s a worthy first effort, and its creators hope to grow the list of tools to include other open source (Drill, Stinger) and commercial6 systems. As these query engines mature and features get added, data from this benchmark can provide a quick synopsis of performance improvements over time.
The initial release includes Redshift, Hive, Impala, and Shark (Hive, Impala, Shark were configured to run on AWS). Hive 0.10 and the most recent versions7 of Impala and Shark were used (Hive 0.11 was released in mid-May and has not yet been included). Data came from Intel’s Hadoop Benchmark Suite and CommonCrawl. In the case of Hive/Impala/Shark, data was stored in compressed SequenceFile format using CDH 4.2.0.
At least for the queries included in the benchmark, Redshift is about 2-3x faster than Shark/on-disk, and 0.3-2x faster than Shark/in-memory. Given that it’s built on top of a general purpose engine (Spark), it’s encouraging that Shark’s performance is within range of MPP8 databases (such as Redshift) that are highly optimized for interactive SQL queries. With new frameworks like Shark and Impala providing speedups comparable to those observed in MPP databases, organizations now have the option of using a single system (Hadoop/Spark) instead of two (Hadoop/Spark + MPP database).
Let’s look at some of the results in detail:
Shark is 100X faster than Hive for SQL, and 100X faster than Hadoop for machine-learning
Hadoop’s strength is in batch processing, MapReduce isn’t particularly suited for interactive/adhoc queries. Real-time1 SQL queries (on Hadoop data) are usually performed using custom connectors to MPP databases. In practice this means having connectors between separate Hadoop and database clusters. Over the last few months a number of systems that provide fast SQL access within Hadoop clusters have garnered attention. Connectors between Hadoop and fast MPP database clusters are not going away, but there is growing interest in moving many interactive SQL tasks into systems that coexist on the same cluster with Hadoop.
Having a Hadoop cluster support fast/interactive SQL queries dates back a few years to HadoopDB, an open source project out of Yale. The creators of HadoopDB have since started a commercial software company (Hadapt) to build a system that unites Hadoop/MapReduce and SQL. In Hadapt, a (Postgres) database is placed in nodes of a Hadoop cluster, resulting in a system2 that can use MapReduce, SQL, and search (Solr). Now on version 2.0, Hadapt is a fault-tolerant system that comes with analytic functions (HDK) that one can use via SQL. Read more…
Cloudera ventures into real-time queries with Impala, data centers are the new landfill, and Jesper Andersen looks at the relationship between art and data.
Here are a few stories from the data space that caught my attention this week.
Cloudera’s Impala takes Hadoop queries into real-time
Cloudera ventured into real-time Hadoop querying this week, opening up its Impala software platform. As Derrick Harris reports at GigaOm, Impala — an SQL query engine — doesn’t rely on MapReduce, making it faster than tools such as Hive. Cloudera estimates its queries run 10 times faster than Hive, and Charles Zedlewski, Cloudera’s cloud VP of products, told Harris that “small queries can run in less than a second.”
Harris notes that Zedlewski pointed out that Impala wasn’t designed to replace business intelligence (BI) tools, and that “Cloudera isn’t interested in selling BI or other analytic applications.” Rather, Impala serves as the execution engine, still relying on software from Cloudera partners — Zedlewski told Harris, “We’re sticking to our knitting as a platform vendor.”
Joab Jackson at PC World reports that “[e]ventually, Impala will be the basis of a Cloudera commercial offering, called the Cloudera Enterprise RTQ (Real-Time Query), though the company has not specified a release date.”
Impala has plenty of competition on this playing field, which Harris also covers, and he notes the significance of all the recent Hadoop innovation:
“I can’t underscore enough how critical all of this innovation is for Hadoop, which in order to add substance to its unparalleled hype needed to become far more useful to far more users. But the sudden shift from Hadoop as a batch-processing engine built on MapReduce into an ad hoc SQL querying engine might leave industry analysts and even Hadoop users scratching their heads.”
You can read more from Harris’ piece here and Jackson’s piece here. Wired also has an interesting piece on Impala, covering the Google F1 database upon which it is based and the Googler Cloudera hired away to help build it.
(Cloudera CEO Mike Olson discussed Impala, Hadoop and the importance of real-time at this week’s Strata Conference + Hadoop World.)