ETERNUS6000 New Models - World's Highest Performance Accelerates Your Business
October 28, 2004
In 2003 the total amount of data being stored by business was identified as in the order of 61 peta bytes. The substantial increases in data storage are said to continue at an annual rate of 59%. Storage systems have become essential devices of corporate IT systems and as well as their ability to store such huge data volumes now must also meet and manage many additional demands. These include consistency of access, high reliability, and performance that ensures stable operation even under peak transaction loads. Fujitsu's ETERNUS6000 storage systems provide optimum solutions that accommodate all of these requirements.
Since their release in October 2003, ETERNUS6000 storage systems have been shipping steadily, boosted by their availability for use with UNIX servers, Intel architecture servers and mainframes. Already highly evaluated, from October 27th 2004 the ETERNUS6000 lineup is further refreshed with a new release of products. These feature even greater power and flexibility to better address and respond to the rapid change in IT environments and business data storage requirements.
ETERNUS6000 Features
The Industry's Highest Performance
Top-of-the-range ETERNUS6000 model 1100 is now powered by 3.2GHz high-speed processors and supports up to 128GB cache memory. In addition, all back-end modules have been enhanced to enable even faster internal processing. The combined result of these comprehensive performance enhancements is the world's highest-level performance in the SPC Bench mark-1™ - the standard by which storage systems are measured. All other models in the ETERNUS6000 range also have outstanding performance with a 40% improvement over previous models.
ETERNUS6000 Model 1100 Achieves World's Record in SPC Benchmark-1™
World's First 146GB/15,000rpm Disk Drive Support
For ultimate performance ETERNUS6000 also supports 146GB/15,000rpm disk drives. This is an industry first and combines cost effectiveness with ultra-high performance. Now you can enjoy maximum configuration flexibility with a wide range of drive choices both in capacity and rotational speed: 73GB, 146GB and 300GB drives at 10,000rpm, and 36GB, 73GB and 146GB at 15,000rpm.
iSCSI Support Enables Storage Integration in IP Network Environments
ETERNUS6000 now provides iSCSI support for use in open system environments. This brings even better cost-effective storage integration alternatives, by enabling the use of your existing IP-network devices.
ETERNUS6000 New Models - Lineup
Top-of-the-range ETERNUS6000 model 1100 provides both 306TB maximum capacity and the world's highest performance. This means ETERNUS6000 can offer its advantages to both business and science and technology. Even calculations requiring state-of-the-art ultra-high capacity and performance can be supported. Models 500, 700 and 900 have also increased their performances by up to 40% compared with previous models. From this new lineup, you will be able to choose the optimum storage system for your business and system requirements.
High-Reliability and High-Performance Technologies
The new ETERNUS6000 supercedes the advantages of previous models in both architecture and functionality.
Internal Processing Architecture for High Throughput
ETERNUS6000 is equipped with four Giga-Byte routers. These are used for internal processing between Controller Modules (CMs), Channel Adapters (CAs), and Device Adapters (DAs). The result is very high total throughput at 17.1GB/sec. Each Giga-Byte router transfers data at 4.27GB/sec. This high level of throughput ensures that processing speed and the highest levels of data reliability are maintained while server data is duplicated and written to the redundant caches.
New Technology to Support Continuous Operation - FC Fabric Architecture
While ETERNUS6000 employs FC fabric (*Note 1) architecture, other vendors typically employ FC-AL (*Note 2) technology. This is used for the connections between Device Adapters on the controllers and disk drives. ETERNUS6000's use of FC fabric architecture overcomes problems associated with FC-AL, where disk failures cause loss of redundancy due to the use of looped connections. With FC-AL reliability and performance are both reduced even if operation can continue.
ETERNUS6000 solves this problem by employing FC fabric architecture. In the FC fabric architecture, duplicate special switches, a unique Fujitsu development, are allocated between the DAs and disk drives. On receipt of an access command from the server, the DA and the target disk drive are directly connected by this switching method.
This FC fabric architecture prevents any single disk drive failure from affecting other related paths. It also maintains the path redundancy and enables continuous access to all other operational drives. This provides you with the best reliability and most stable performance possible.
*Note 1: FC-fabric: A Fibre Channel network topology. Here Device Adapters (DAs) and disk drives are connected via one or
more Fibre Channel switches. This topology works as if DAs and disk drives are directly connected in a one-to-one relationship.
*Note 2: FC-AL (Fibre Channel-Arbitrated Loop): Another Fibre Channel network topology. Here Device Adapters (DAs) and disk
drives are connected in the form of a loop. So direct connection in a one-to-one relationship is not possible.
Further Reliability Enhancements
All major components of ETERNUS6000 storage systems, including disk drives, power supply units, fans and batteries are redundantly equipped and hot-swappable for high reliability. Up to four controllers can be installed. With such redundancy of the configuration, the impact in the unlikely event of a controller failure is localized, so that other parts of the storage system are not affected by the failure. Importantly system redundancy is maintained ensuring stable continuous operation.
As shown in the figure below, ETERNUS6000 RAID-member drives can be allocated across different Drive Enclosures (DEs). This provides continuous access to data even when one DE fails and all the disks in the DE become inaccessible. This feature ensures your business is not interrupted.
High-reliability testing technology
High storage systems reliability means employing highly reliable components. However it is impossible to ensure no occurrence of a hardware failure.
High reliability goes beyond the hardware redundancy described above, and includes the important provision that the storage system can detect, isolate faulty parts, and properly recover itself. Fujitsu tests every ETERNUS6000 component's recovery capability at the design stage. This includes simulating hardware failures in all parts using an auto-clip-testing device (*Note 3) that verifies recovery functions work properly.
In the same way, simulated failures are generated for disk drives to verify their firmware recovery functions.
Fujitsu's extensive use of testing technology enables the most obscure fault circumstances to be observed enabling the production of the most highly reliable storage systems.
*Note 3: Auto-clip-testing device: a device that generates simulated failures in component circuits by electrically short-circuiting them.
Summary
For IT systems requiring 24 x 7 x 365 non-stop operation, ETERNUS6000 storage systems support stable operation, provide high system availability, enable data assurance, and enhance operational efficiency in cooperation with today's cutting-edge software.
For more information on ETERNUS6000 features and solutions, please refer to the following pages.
Related information:
- Remote mirroring and fast recovery - Remote Advanced Copy/Extended Remote Advanced Copy
- Advanced management stable operation for any type of storage system, including SAN, DAS and NAS
