MemSQL Delivers Top Performance on Industry Benchmarks

Becoming the first cloud-native database to pass all three industry benchmarks for analytics and transactions, meeting or beating all published results

San Francisco, California, UNITED STATES

SAN FRANCISCO, May 14, 2019 (GLOBE NEWSWIRE) -- MemSQL, the No-Limits Database™ for operational analytics and cloud-native applications, today unveiled new benchmark results that prove it’s the ideal database for the broadest set of workloads. MemSQL delivered best-in-class performance, at a dramatically lower cost, all on standard cloud infrastructure, across the top three industry-standard benchmarks for transactional and analytical workloads.

Piecing together data infrastructure slows down time to insight and action, and with these benchmarks MemSQL shows that buying specialized data stores for each new workload is a thing of the past. MemSQL now proves you can run high performance, ACID-compliant transactions and complex analytics on the same system, eliminating architecture complexity while reducing costs.

MemSQL Delivers Top Performance on Industry Benchmarks - Click to Tweet

The need to stay competitive is creating more pressure on businesses to deliver insights and make faster decisions on increasing volumes of data. Technology teams at these companies need a mature data infrastructure that can meet a broad set of requirements, perform well at scale, and fit easily with existing processes and tools. However, the database market is fragmented and offers dozens of specialized databases that do one thing well but most other things poorly. This creates data silos within the enterprise as they spin up one specialized database after another, burdening organizations with extra cost and complexity.

MemSQL is on a mission to change all that. It has built a fully integrated, general-purpose SQL database that simplifies data infrastructure by delivering support for multiple workloads and data types while being the fastest database in real world conditions and at the lowest cost for performance.

“Traditionally, organizations could not perform analytics on data stored in operational transactional systems without setting up a separate database for this purpose,” said James Curtis, senior analyst, Data, AI & Analytics Data at 451 Research. “With the blending of operational and analytical systems, this is no longer necessary, turning the achievement of high performance into a reality for many enterprises.”

The Benchmark Results
To highlight the breadth of its capabilities, MemSQL used three common benchmarks that are typically used separately to show off the distinctive performance capabilities of database and data warehouse vendors. MemSQL successfully ran all three benchmarks, the first modern database company to do so, with performance being up to 8 times faster than competitors.

The Transaction Processing Performance Council (TPC) is a compilation of 14 industry-standard benchmarks used to evaluate the performance of different technologies. MemSQL ran (uncertified but reproducible) TPC-C, TPC-DS, and TPC-H, to become the first modern database vendor to complete all three benchmarks and show that it’s the only vendor capable of meeting or beating a wide range of vendors in the industry. Making MemSQL the only cloud-native, distributed database that can run both analytical and transactional workloads in one system.

  • TPC-C measures how fast transactions can be executed against a database. The transactions being simulated are order entry - transactions that are representative of activity that occurs with any industry that manages, sells, or distributes products or services. Simply put, this test shows the degree to which a transaction database can scale. In this benchmark, MemSQL was up to 8 times faster than the competition.
  • TPC-DS measures the performance of decision support systems, which examines large volumes of data, gives answers to business questions, and executes on ad-hoc queries. MemSQL’s performance against the competition varied on this benchmark but was found to take only about a third as long as some vendors.
  • TPC-H also analyzes decision support systems that examine large volumes of data and execute queries with a high degree of complexity, but also gives answers to critical business questions. MemSQL’s performance ranged from similar, to as much as two times faster than the competition.

MemSQL customers get additional performance gains from the ability to run fast ingest, transactions, and analytical queries on a single database, with high concurrency support, and all without the need for extract, transform, and load (widely known as ETL) operations and other time-consuming overhead.

“We’ve seen this performance firsthand,” said John Toups, CTO at Medaxion, a software company improving productivity and quality for anesthesia practices. “With our old database it took 20 to 30 minutes to run analyses on our data; with MemSQL it takes under a minute. Having a highly performant database allows us to deliver real-time insights to our customers.”

More details on the benchmark and the results are available on the MemSQL blog.

Evolved Enterprise Features
Also announced is the latest version of the product, MemSQL 6.8, which extends the features previously released in MemSQL 6.7 and builds on the database’s analytical abilities. With this latest version of MemSQL, we’re announcing advanced security support with Kerberos and wire encryption for HDFS Pipelines along with innovative ad hoc query performance boosts. For customers that demand the strictest requirements in information security, these additions give confidence that end users achieve optimal query response times without sacrificing proper authentication and full data protection during transport.

With these new features, even more Hadoop implementations can be accelerated and simplified with MemSQL. Additional new features to MemSQL 6.8 include:

  • Faster ad-hoc queries – We added just-in-time compilation, what we’re calling Interpret First Query Compilation, to MemSQL, which allows customers to not have to go through the entire compilation step before starting to run code. This new feature allows for first run queries to run 10-100 times faster, without hints or tuning.
  • More SQL goodness – MemSQL 6.8 has expanded SQL coverage to simplify query design and ease migrations from legacy databases supporting ANSI SQL. New queries include RIGHT SEMI, ANTI, and OUTER JOIN for added performance and scale.

“No one wants to store data in multiple places or increase the cost and complexity of doing business. But that’s exactly what today’s special-purpose databases are forcing customers to do,” says MemSQL CEO Nikita Shamgunov. “That’s why we are taking a different approach. We have spent the last six years developing the world’s fastest database, and we can now assure customers that it is the best database for the vast majority of workloads.”

MemSQL 6.8 is available immediately. To get started for free, try MemSQL today.

About MemSQL
MemSQL is The No-Limits Database™, powering modern applications and analytical systems with a cloud-native, massively scalable architecture for maximum ingest and query performance at the highest concurrency. MemSQL envisions a world where every business can make decisions in real time and every experience is optimized through data. Global enterprises use the MemSQL distributed database to easily ingest, process, analyze, and act on data in order to thrive in today’s insight-driven economy. MemSQL is optimized to run on any public cloud or on-premises with commodity hardware. Visit or follow us @memsql.


Contact Data