Printer Friendly

PYRAMID SHIPS VALENCE CLUSTER SOLUTION TO SUPPORT ORACLE7 PARALLEL SERVER

 SAN JOSE, Calif., April 19 /PRNewswire/ -- Pyramid Technology Corporation (NASDAQ: PYRD) today announced general availability of Valence(TM) Cluster, its clustered configuration of loosely coupled mainframe-class Pyramid MIServer(R) systems (2, 4 and 12 S or ES servers) which support Oracle Corporation's ORACLE7 Cooperative Server with the Parallel Server option. All MIServer nodes within the cluster dynamically share the ORACLE database to provide continuous database availability, a crucial need for data centers running mission-critical applications.
 By employing redundancy, Valence Cluster protects users from single points of failure, whether the failure is hardware, software or network- related. As long as one node in Valence Cluster remains operational, the ORACLE database remains accessible, avoiding costly downtime. Unlike other solutions, with Valence Cluster, all computing resources perform productive work to maximize performance, bandwidth and capacity. Valence Cluster achieves continuous database availability by combining the computing resources offered by each MIServer node without compromising the capacity and throughput demands of the data center.
 Supporting up to four MIServer nodes, Valence Cluster also provides superior scalability to data centers running very large databases or supporting very large user communities. While Pyramid's current symmetric multi-processing (SMP) implementation offers near linear scalability of up to 24 processors for each MIServer, the scalability of the clustered nodes greatly exceeds single SMP system boundaries, supporting larger user communities and expanding database capacity. For example, a fully configured Pyramid Valence Cluster configuration can support 4 GB of random access memory and 96 CPUs with disk storage in excess of 250 gigabytes.
 Valence Cluster also greatly simplifies system and database administration by consolidating computing and database resources into a single logical resource. Valence Cluster also permits single-point database and cluster administration, with all management capable of being conducted from the same console.
 "Pyramid's Valence Cluster running the ORACLE7 Parallel Server attracts a broad range of data center customers looking for robust and cost-effective database computing solutions capable of offloading the mainframe," said John Chen, chief operating officer for Pyramid. "Prime candidates for Valence include customers deploying a single MIServer with increasing data availability requirements; data centers acquiring Oracle applications looking for an open system solution; large database sites with user communities exceeding existing operating system and memory resources; and organizations seeking a combination of availability, capacity and consolidated database resources."
 "ORACLE7 with Parallel Server option provides the software reliability and scalability required by mission-critical applications," said James Sha, vice president of Oracle's UNIX Products Division. "Pyramid's robust cluster technology and its unique Fault-Tolerant Distributed Lock Manager provide a solid systems base for ORACLE7 Parallel Server. Our joint customers can expect continuous database availability and scalability beyond the boundaries of traditional SMP architecture."
 Among the early evaluators of Valence Cluster is the Toronto Stock Exchange (TSE), Canada's largest stock exchange, which is transitioning from a traditional trading floor to a paperless computerized trading system.
 Having served as an alpha site for ORACLE7 Cooperative Server, TSE chose Pyramid's Valence Cluster as the development environment for its Oracle order trading database.
 According to Wayne Fowler, director of technology and network services at the Toronto Stock Exchange, "We are committed to downsizing to an open systems implementation and Pyramid will play a key role in that strategy. Pyramid's relationship with Oracle and their implementation of ORACLE7 Cooperative Server with the Parallel Server option have been important factors in our decision."
 Key Valence Cluster Differentiators
 "Pyramid's engineering expertise gives Valence Cluster the promise of providing one of the better solutions supporting the ORACLE7 Cooperative Server with the Parallel Server option," said John Morrell, UNIX Service analyst at International Data Corporation. "We believe the Valence Cluster solution, unique in its implementation of Virtual Ethernet, Fault-Tolerant Distributed Lock Manager, and centralized system management, provides a real added-value customer solution, giving the company a significant competitive edge."
 Valence Cluster maintains the following critical technology differentiators that provide continuous database availability and maximum utilization of computing resources:
 -- Fault-Tolerant Distributed Lock Manager: Pyramid's Fault- Tolerant Distributed Lock Manager, which is integrated into the System V Release 4 kernel, coordinates concurrent access to the shared ORACLE database and facilitates Oracle's ability to minimize contention. The Distributed Lock Manager is fault-tolerant because it is distributed on every node in the cluster. In the event of a failure that disrupts the service of a lock manager on one node, lock managers on operational cluster nodes continue to perform. The Distributed Lock Manager software is engineered to prevent failures from cascading from one node to another. When the failed node is recovered and the Distributed Lock Manager is reinstated, it uses the other nodes in the cluster to restore its state as the node regains its logical connection with the shared database.
 -- Virtual Ethernet: Pyramid is the first and only vendor to implement Virtual Ethernet for an ORACLE7 Parallel Server solution. Virtual Ethernet is a layer of software residing on an MIServer that manages multiple Ethernet connections as a single logical connection which results in greatly enhanced performance. Rather than sending mirrored communications across the physically redundant connections, Virtual Ethernet actively utilizes both physical channels by means of load balancing and data striping. As a result, bandwidth is increased when multiple connections are fully functional which enables Pyramid's implementation to maintain the highest degree of throughput.
 -- Node Interconnection: Valence Cluster's unique node interconnection combines high-speed, low-latency and physical redundancy without compromising stringent adherence to standards. The interconnection between cluster nodes is based on a private network. Consequently, there is no unnecessary traffic to impede communication speed and response. In addition, a new Pyramid-designed Ethernet controller supports high-performance packet transfer between nodes. Pyramid's implementation of the low latency protocol for Ethernet minimizes latency and guarantees message delivery between nodes. The Valence Cluster also provides physically redundant channels of communication, all of which are fully utilized to increase bandwidth. Pyramid's implementation of Ethernet communications adheres to all required standards.
 -- High Availability Array System (HAAS): Pryamid's HAAS is a highly available mass storage SCSI-II I/O subsystem specifically designed to support multi-host environments. Because an I/O subsystem can limit the availability of the overall cluster, Pyramid has designated HAAS as a Valence Cluster requirement. HAAS maintains high data availability levels by ensuring that no single hardware failure will disrupt data access. For example, HAAS is equipped with redundant power supplies, cooling systems and SCSI interfaces. The subsystem also supports distributed RAID-1 mirror disk array.
 -- Distributed Mirror Disk: Pyramid Valence Cluster implements a Distributed Mirror Disk (distributed RAID-1) to eliminate the possibility that media failure would disrupt data access. Within the I/O subsystem, the Distributed Mirror Disk communicates with its counterparts on other nodes within the cluster to maintain consistency. In the event of a device failure, the Distributed Mirror Disk utilizes the Distributed Lock Manager to interact with the ORACLE database to re- establish the mirror.
 -- Centralized Cluster Management: To facilitate the complexities of loosely coupled environments, Pyramid provides a single console for centralized clustered management. The single console is an intelligent workstation running SVR4 UNIX.
 -- DC/OSx: Pyramid's DC/OSx operating system is a robust version of System V Release 4 UNIX specifically designed for very large data center environments, and is ideally suited to meet the requirements of downsizing customers deploying mainframe-class solutions with Pyramid systems.
 Availability and Pricing
 The Valence Cluster is now generally available, with pricing dependent upon each customer's particular MIServer configuration. The enabling Valence software is priced at $200 per named user and $300 per concurrent user.
 Company Background
 Pyramid Technology Corporation manufactures, markets and supports the MIServer line of high-performance, high-availability servers for open data center computing. Pyramid's products are based on a multiprocessing RISC system architecture and a commercially enhanced version of UNIX System V Release 4 (SVR4) operating system. Pyramid has installed over 3,500 systems internationally, and supports the largest UNIX-based relational databases in the world. The company was founded in 1981 and is headquartered in San Jose.
 -0- 4/19/93
 /CONTACT: Stuart McFaul of Pyramid Technology, 408-428-8460/
 (PYRD)


CO: Pyramid Technology Corporation ST: California IN: CPR SU: PDT

CK -- NY012 -- 7163 04/19/93 08:33 EDT
COPYRIGHT 1993 PR Newswire Association LLC
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1993 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Publication:PR Newswire
Date:Apr 19, 1993
Words:1343
Previous Article:HOMEDCO ANNOUNCES PRELIMINARY 2ND QUARTER OPERATING RESULTS
Next Article:PYRAMID AND STRATEGIC PARTNERS LAUNCH 'AFFILIATE SUPPORT' SERVICE FOR MISSION-CRITICAL DATABASE ENVIRONMENTS
Topics:


Related Articles
PYRAMID ANNOUNCES DATA CENTER CLUSTER SOLUTION WITH ORACLE7 PARALLEL SERVER
PYRAMID TECHNOLOGY REPORTS REVENUE AND EARNINGS INCREASES
PYRAMID ANNOUNCES CORPORATE ROADMAP FOR ENTERPRISE CLIENT/SERVER COMPUTING
PYRAMID ANNOUNCES IMMEDIATE AVAILABILITY OF ORACLE7 RELEASE 7.1 PARALLEL DATA MANAGEMENT SOFTWARE
ORACLE DEMONSTRATES FIRST SCALABLE CLUSTER SOLUTION ON PENTIUM PRO HIGH-VOLUME INTEL-BASED SERVERS
PYRAMID AND SIEMENS NIXDORF ANNOUNCE SUPPORT FOR ORACLE UNIVERSAL SERVER
Oracle Solution Running SCO UnixWare Involves Compaq, CLAIRiion, Digital Equipment, IBM, Intel and Unisys
Pyramid and Siemens Nixdorf Announce Special Release Availability of Sap's R/3 Version 3.0D and Oracle7 Release 7.3.2 with Parallel Server Option on...
Oracle Announces Oracle7 Parallel Server Release 7.3 for Sun Ultra(TM) Enterprise Cluster PDB Family
Oracle Announces Record Database Benchmark Result on IBM RS/6000 Clustered SMP Platform

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters