Client/Server Labs develops first large scale benchmark for the database server.ATLANTA--(BUSINESS WIRE)--March 4, 1996--Client/Server Labs, the owner and manager of RPMark95, announces the upcoming release of its new benchmark, RPM/dbs. This new client/server benchmark, which focuses on the overall balanced performance of mid-size to enterprise-class servers, is the first open, scaleable, mixed workload benchmark of its kind. Due out at the beginning of the second quarter, RPM/dbs will offer end users and vendors an affordable, comprehensive, and objective measurement of large client/server solutions. "The development of RPM/dbs is in response to end users needing a mixed-workload test which measures sizable high-end client/server systems, systems which can support hundreds of users," said Bill Wallet, vice president of Client/Server Labs. "With this new benchmark, end users will have the ability to test and measure the performance and feasibility of any type or size client/server configuration they are contemplating." "After a well-regarded start with the RPMark95 benchmark last year, Client/Server Labs is responding with RPM/dbs to a real market need for useful, reliable information about high-end, mixed-workload commercial performance," said Peter Kastner, vice president of the Aberdeen Group Aberdeen Group is a provider of business-related research services. It has its headquarters in Boston, Massachusetts and belongs to the Harte-Hanks group. Founded in 1988, Aberdeen's research is used by over 2. , an industry consulting and market research company. WHAT THE RPM/dbs MEASURES Like the RPMark95 benchmark, RPM/dbs can be used to test any aspect of a client/server system. It can measure the impact of a single component, such as the server, database software, operating system operating system (OS) Software that controls the operation of a computer, directs the input and output of data, keeps track of files, and controls the processing of computer programs. , drivers, network application, etc. There are no hardware or software limitations for running the benchmark, making it extremely portable across a variety of hardware architectures, operating systems Operating systems can be categorized by technology, ownership, licensing, working state, usage, and by many other characteristics. In practice, many of these groupings may overlap. and databases. For example, if one were considering two types of database software to support hundreds of users and wanted to determine which one outperformed the other under a proposed configuration, then the RPM/dbs could measure this. RPM/dbs is comprised of multiple workloads running simultaneously on a server configuration. They are: --On-Line Transaction Processing Updating the appropriate database records as soon as a transaction (order, payment, etc.) is entered into the computer. It may also imply that confirmations are sent at the same time. Transaction processing systems are the backbone of an organization because they update constantly. (OLTP (OnLine Transaction Processing) See transaction processing and OLCP. OLTP - On-Line Transaction Processing ), this component is based on a highly modified subset of the TPC-C A benchmark that measures overall transaction processing performance. See TPC. benchmark. --Decision Support (DSS (1) (Digital Signature Standard) A National Security Administration standard for authenticating an electronic message. See RSA and digital signature. (2) (Digital Satellite S ), this component is based on a highly modified subset of the TPC-D A benchmark that measures decision support performance. See TPC. benchmark. Here multiple PC clients are used to initiate query-intensive SQL SQL in full Structured Query Language. Computer programming language used for retrieving records or parts of records in databases and performing various calculations before displaying the results. calls on the server. RPM/dbs also offers end users scaleability. The benchmark is currently configured to represent the workload of approximately 450 "real world" users, yet this number can be expanded upon. Presently, RPM/dbs' OLTP transactions are concentrated in a 1.2 GB database and DSS transactions are concentrated in a 10GB database. The RPM/dbs benchmark measures performance and price/performance. The performance score is based on the difference between the performance of the configuration being tested and the benchmark's hypothetical reference platform. The higher the rating the better the performance. The price/performance is calculated by dividing the total cost of the tested system by the RPM/dbs score. The cost of running the RPM/dbs benchmark begins at $25,000. Tests are administered and certified on-site at Client/Server Labs. Vendors may also purchase an annual site license for $2,500 to use the RPM/dbs test suite in development. Testing reports are available through Client/Server Labs and all results of certified tests are listed on Client/Server's web page at www.cslinc.com on the world wide web. Client/Server Labs offers a full suite of hardware and software testing services in addition to the RPM/dbs and the RPMark95. The company is located at 8601 Dunwoody Place, Suite 332, Atlanta, GA 30350-2509. For more information on Client/Server Labs, call 770/552-3645 and 770/993-4667 (fax), or visit our web site. -0- EDITOR'S NOTE: Client/Server Labs and RPMark95 are registered trademarks of Client/Server Labs, LLC (Logical Link Control) See "LANs" under data link protocol. LLC - Logical Link Control . CONTACT: Client/Server Labs Elizabeth W. Clarke, Director of Public Relations public relations, activities and policies used to create public interest in a person, idea, product, institution, or business establishment. By its nature, public relations is devoted to serving particular interests by presenting them to the public in the most 770/552-3645 eclarke@cslinc.com |
|
||||||||||||||
Printer friendly
Cite/link
Email
Feedback
Reader Opinion