Quantcast
  • E-mail
  • Print
  • Comment
  • Font Size
  • Digg
  • del.icio.us
  • Discuss article

The ParAccel Analytic Database Running on Sun Servers Shatters 100 Gigabyte, 300 Gigabyte, and One Terabyte TPC-H Benchmark Records for Performance and Price/Performance

Posted on: Monday, 29 October 2007, 09:00 CDT

ParAccel, Inc. a provider of high-speed, columnar database solutions for all forms of analytic processing, announced today that the ParAccel Analytic Database running on Sun Fire™ X4100 servers set TPC-H (ad-hoc-decision support) benchmark world records for performance and price/performance at the 100 gigabyte, 300 gigabyte, and one terabyte database sizes. The TPC-H ranks top ten results for performance and price/performance across six database sizes; Sun and ParAccel solutions are now the proven performance and price/performance leaders of the first three.

At the 1TB database size, a Sun-ParAccel system delivered performance of 315,842 Composite Queries per Hour (QphH) and price performance of $4.57. These impressive results topped the prior performance leader by more than 4X at about half the cost. That same system beat the previous price/performance leader by nearly 2X, while delivering more than 56X its performance. Furthermore, this same Sun-ParAccel system loaded 1TB of data in only 41 minutes versus the prior performance leader's nearly 6 hours, and used only 13TB total storage versus that system's 40TB. Similar performance gains were recorded at the 100 and 300 gigabyte database sizes.

"Our database is a scalable, analytic super computer at an affordable price," said Barry Zane, CTO of ParAccel. "Using a car analogy, we're giving you a Ferrari for the price of a BMW with twice the fuel efficiency of a Prius."

ParAccel also announced it is working with Sun to deliver extremely high performance yet cost-effective data warehousing solutions for Sun's server platforms marketed as the Ultra-Fast Data Warehouse Powered by Sun and ParAccel. The announcements were made at the TDWI World Conference where the solutions are on display.

The Ultra-Fast Data Warehouse is delivered as a pre-configured, pre-staged system that is ready to load and operate. The Ultra-Fast Data Warehouse is available in two configurations: Sedona, a storage-based solution, and Phoenix, a memory-based solution. Both solutions combine the small footprint, scalability and power efficiency of the Sun Fire X4100, Sun Fire X4500 or Sun Blade 6000 Modular Systems and take advantage of ParAccel's columnar orientation and unique adaptive data compression techniques.

"The Ultra-Fast Data Warehouse powered by Sun and ParAccel analyzes terabytes of information at record breaking speeds and has established performance records on cost effective, energy efficient Sun platforms," said Lisa Sieker, Vice President of Systems Marketing at Sun Microsystems. "ParAccel's offering fits well with Sun's strategy to provide eco-friendly, high performance business solutions that help reduce cost."

"We are delighted to be working with Sun to deliver the world's fastest and most efficient platform for analytic queries," said ParAccel President and CEO David Ehrlich. "Our audited benchmark results confirm that the Sun-ParAccel solution is not only the world's fastest at numerous data sizes, but also the world's most cost-effective. We anticipate even more great news ahead from our commitment to Sun's system offerings. For example, ParAccel's upcoming release on the Solaris™ operating environment should show even more impressive performance results. Our combined innovation power is providing customers with a much needed breakthrough in analytic query performance."

According to Richard Winter, president of Winter Corp, experts in large scale data management, "These are impressive benchmark results, especially when you take into account that they utilize a single configuration at each scale factor to top both the performance and price/performance lists," Winter said. "And the margin of leadership is not small: ParAccel has more than four times the performance and nearly twice the price/performance of the next best results."

About the TPC-H Benchmark

The TPC Benchmark™H (TPC-H) is a decision support benchmark. It consists of a suite of business oriented ad-hoc queries and concurrent data modifications. The queries and the data populating the database have been chosen to have broad industry-wide relevance. This benchmark illustrates decision support systems that examine large volumes of data, execute queries with a high degree of complexity, and give answers to critical business questions.

The TPC-H benchmark represents decision support environments where users don't know which queries will be executed against a database system; hence, the "ad-hoc" label. Pre-knowledge of the queries may not be used to optimize the DBMS system. Consequently, query execution times can be very long.

To review details regarding the ParAccel-Sun TPC-H benchmark results, please visit http://www.tpc.org/tpch/.

About ParAccel

ParAccel, Inc. is the proven leader in analytic DBMS performance and price/performance. The ParAccel Analytic Database combines patent-pending innovations with a columnar, compressed, massively parallel, all-in-memory-capable architecture and is available as a virtual or packaged appliance on standard hardware from all major vendors. It can be implemented stand-alone, or as a drop-in accelerator to extend an existing SQL Server or Oracle investment. ParAccel's management team includes technical founders and industry veterans from noted data management companies Netezza, Oracle, Teradata, Gupta and PointBase. ParAccel recently shattered the TPC-H benchmark performance and price/performance records at 100GB, 300GB and 1TB using a single configuration at each scale factor. ParAccel is headquartered in San Diego, CA. For more information please contact us at info@paraccel.com or 866-903-0335 X122, or visit us at www.paraccel.com.

©2007. All rights reserved.


Source: Business Wire

More News in this Category


Related Articles



Rating: 3.0 / 5 (10 votes)
Rate this article:
1/52/53/54/55/5

User Comments (0)

Comment on this article

Your Name
Text from the image
Comment
max 1200 chars
* All fields are required