Read PREMA Architecture for the ProLiant DL980 G7 server - Overview text version

HP PREMA Architecture for ProLiant DL980 G7 server


The HP PREMA Architecture is the design foundation for scale-up x86 servers that need to deliver more scalability, resiliency, and efficiency. This technology addresses the needs of customers using x86 servers for the most demanding, data intensive workloads such as business processing, business intelligence, and large scale consolidation and virtualization. HP scale-up x86 servers utilizing the HP PREMA Architecture are designed to take full advantage of processor technology such as Intel® Xeon® processor 7500/6500 series with Quick Path Interconnect (QPI) and the industry standard operating environments such as Microsoft ® Windows® Server and Linux. HP ProLiant DL980 G7 extends the performance and resiliency enhancements provided by the underlying processor technology and operating environment with the HP PREMA Architecture through a node controller design. When a system scales to a larger number of inter-connected processors, the communication and coordination between the processors grows at an accelerated (exponential) rate creating a system bottleneck. The node-controller relieves this bottleneck with two key features, smart CPU caching and the redundant system fabric. The result of this design is both a reduction to this communication overhead between the processors and an enhancement to system resiliency

Balanced scaling with smart CPU caching

Smart CPU caching groups the 8 processors of the DL980 G7 server into "islands" of two. A node controller then interfaces with each "island" to reduce inter-processor communication and to enable rapid access to local memory. Processors in the "island" have access to the cache state information about all the processors decreasing the overhead of requesting updates from all other processors. The ProLiant DL980 G7 server offers up to 20% better system performance than competitive 8 socket systems for a diverse set of workloads.* With up to 30% faster local memory access,* workloads with affinity to local memory and the applications isolated to a processor or an "island" can gain additional performance

advantages. For example, virtual machines, SAP, and high performance computing applications can experience performance gains on a DL980 G7. Some of the key benefits include:

Dynamic Routing--Improves system reliability and performance by dynamically rerouting traffic through an alternate path in the event of an overloaded link condition.

Minimized inter-processor communication--Reduces Rapid recovery--Improved error logging and the overhead of requesting and receiving updates from diagnostics information enables the operating system all other processors to improve system performance. or hypervisor to attempt error recovery or enables Accelerated local memory access--Frees up memory the system administrator to take corrective actions. The ProLiant DL980 warm resets after fatal errors and bandwidth from the burden of memory tag updates captures error logs instead of a hard reset and losing for certain remote IO traffic. the error log data. Dynamic routing of traffic--Decreases performance bottlenecks by dynamically routing the excess traffic Breakthrough efficiencies from through an alternate path in case a fabric link is ProLiant innovations overused by a processor. In addition to the balanced scaling and self-healing Self-healing resiliency via redundant resiliency enabled by the HP PREMA Architecture, the HP ProLiant DL980 G7, with help from key ProLiant system fabric innovations like iLO3, Insight software and Thermal The redundant system fabric is the interconnect Logic, delivers breakthrough efficiencies. Customers fabric which extends the advanced reliability of the can reduce their data center footprint by consolidating Intel Xeon processor 7500/6500 series to provide 197 legacy servers to a single HP DL980 G7** and higher interconnect bandwidth, lower data error rates realize a payback in as little as 60 days. and dynamic traffic routing for improved resiliency in the ProLiant DL980 G7 server. Some of the key * Based on HP internal engineering benchmarks. features include: ** ProLiant ROI calculator--based on HP internal testing comparing Redundant data paths--System fabric with 50%

hardware on DL360 G4 to DL980 G7

more interconnect links improves system availability by reducing the down time in the event of an interconnect link failure.

To learn more about the HP PREMA Architecture and the ProLiant DL980, go to:

HP ProLiant DL980 G7 server web page

Share with colleagues

Get connected

Get the insider view on tech trends, alerts, and HP solutions for better business outcomes

© Copyright 2010 Hewlett-Packard Development Company, L.P. The information contained herein is subject to change without notice. The only warranties for HP products and services are set forth in the express warranty statements accompanying such products and services. Nothing herein should be construed as constituting an additional warranty. HP shall not be liable for technical or editorial errors or omissions contained herein. Microsoft and Windows are U.S. registered trademarks of Microsoft Corporation. Intel and Xeon are trademarks of Intel Corporation in the U.S. and other countries. 4AA2-5370ENW, Created July 2010


PREMA Architecture for the ProLiant DL980 G7 server - Overview

2 pages

Report File (DMCA)

Our content is added by our users. We aim to remove reported files within 1 working day. Please use this link to notify us:

Report this file as copyright or inappropriate


You might also be interested in

HP DL980 Customer Chronicles
White Paper
HP ProLiant DL980 G7 Server - (US English)