Server: Microsoft-IIS/3.0 Date: Fri, 21 Nov 1997 03:38:53 GMT Content-Type: text/html Accept-Ranges: bytes Last-Modified: Sat, 15 Nov 1997 21:31:15 GMT Content-Length: 24447 Synchronix

SynchronixTM

Chosen by hardware and software industry leaders because they are serious about their data storage.


  • Investment Protection
    Modular design grows with your business.

  • Superior Cache Management System
    Three phase commit of writes.
    Write-back and write-through.
    User Configurable.
    Non-segmented, adaptive.

  • Supports Dual Initiation/Dual Porting
    No system failure on bus failure.

  • Fully Redundant, Hot Swappable Components
    All active components are hot swappable and user replaceable.

  • Concurrent Support of Multiple RAID Levels:
    0, 1, 3, 5, 10, and JBOD

  • Array-based, Automatic Failover
    Failover is automatic and transparent to the host. No drivers required on host.

  • Up to 819GB of scalable storage per cabinet

  • Intuitive, Intelligent Graphical User Interface
Continuously Available Data Storage
Open storage for Open Systems.

Multiple-Host Support Increases Flexibility

By adding a secondary interface to the primary array controller, Synchronix has the ability to configure the unit for either single or multiple host environments.

Multi-hosting to similar systems enables server redundancy failover. If one host fails, another can take control and service applications of the failed server. Users remain active for mission critical applications.

In a multi-host environment- even with servers from different suppliers - the Synchronix Storage Management System allows storage capacity to be shared between systems. By eliminating the need for entirely separate and different storage subsystems, users can benefit from a significant cost savings

Additionally, a common storage server simplifies management and backup for systems administrators who only need to be familiar with a single command set and user interface.

The Synchronix Storage Management System features a high performance, fault tolerant architecture to satisfy large-scale, mission critical applications for open systems (UNIX and NT) environments. Synchronix has been designed to meet growing needs for a storage subsystem with high performance, flexible configurations and investment protection. It provides a scalable, continuously available, feature-rich data storage solution ideally matched for Unix and NT computing environments.

Synchronix has an intuitive Graphical User Interface (GUI), redundant controllers, cross-platform and dual host support making it the perfect fit for today's changing business conditions. A modular design protects your investment by providing flexible growth options to keep pace with your changing business requirements.

Dual Host & Initiation Configuration

Maximum Fault Tolerant Design
All active components, including disk drives, power supplies, fans and system controllers are fully redundant and hot swappable, meaning that the system will remain on-line even in the unlikely event of multiple component failures. If Synchronix detects a power failure in either the power supplies or an external source, it will automatically change caching to a write-through mode and flush all cached data to the disks. The dual controllers automatically monitor and maintain coherency between both sets of cache, allowing one controller to take over if the other one fails. Used with a commercially marketed High Availability Server Failover Software, Synchronix supports multiple host initiation in the event of a server failure.
Exceptional Cache Management Secures Superior Performance
RAID level 5 implementations typically cause severe write performance degradation in many applications that benefit from high availability. However, with ECCS' write-back technique, Synchronix optimizes random read/write operations. Synchronix achieves RAID Level 5 read/write performance that is typically three times faster than other products in its class.

With write-back cache, the write request is acknowledges to the host as soon as the data is in cache and the actual disk update can occur at a later time. Performance greatly improves because the application does not have to wait for write completion.

Synchronix utilizes an exclusive Data Consistency Check. This Data Consistency Check ensures that data is properly written to disk. The non-segmented, adaptive cache is highly user configurable via the GUI.

This superior cache management system allows users to obtain maximum performance without I/O bottlenecks common to other implementations. To further protect against data loss, ECCS offers an intelligent UPS as an option. The UPS interfaces through one of the two Synchronix serial ports. In the event of a power failure, Synchronix can flush the cache to storage and orchestrate a graceful power-down.

With ECCS' write gathering technique, Synchronix can update multiple, continuously stored data blocks on a disk with a single write operation, thereby reducing the number of separate I/O transactions.

BENEFITS OF ECCS WRITE GATHERING TECHNIQUE

Feature Benefit
Fully redundant, Hot Swappable Components All active components including drives, controllers, power supplies and fans are completely redundant and hot swappable. No Down Time.
Supports Multiple RAID Levels 0, 1, 3, 5, 10, JBOD Allows Multiple platforms to run multiple RAID levels on the same system. Also allows the user to place data in the most cost effective array.
Intuitive Graphical User Interface Point and click administration. Simplified status monitoring, configuring, performance analysis and diagnostic control. Manage multiple Synchronix units for a single GUI.
Supports Dual Initiation No system failure on bus failure. Separate SCSI buses for each controller protects against bus failure. System does not stop on bus failure.
Array-based, Transparent Failover Failover is automatic and transparent to the host. No host intervention required. Eliminates critical delays. No drivers required on host.
Data Consistency Check Three phase commit of writes ensures that data is properly written to disk.
Superior Cache Management Unsurpassed performance in its class.
Maximum Fault Tolerance Proactive response. Eliminates downtime.
Scalability Investment Protection. Modular design grows with your business.
Dual Port LUNS can be mapped to multiple servers.

The Synchronix Storage Management System also uses a read ahead optimization feature that enhances sequential read operations. By monitoring data access requests for sequential patterns, Synchronix can begin filling its read cache with data from all drives in a strip, in anticipation of its actual need, thereby minimizing wait times.

 
Performance
The Synchronix controller supports Tagged Command Queuing to the disk drives, with up to 255 Tags. This means that 255 requests can simultaneously be processed over 16 logical disk drive volumes. For RAID 1 and 10 volumes, Synchronix further increases performance by using Inverse MirroringTM to minimize disk drive seek times.

Scalable Features
Each Synchronix Controller supports a maximum of 30 devices (e.g. disk drives) which can be configured in 10, 20, and 30 device enclosures. Three 30 device enclosures can be housed in a single ECCS 70" Data Center Cabinet for a total of 90 devices. The controller is designed to use devices of various capacities, so the subsystem will be able to use higher capacity and better performing drives as they become available.

Synchronix uses 16MB SIMMS for system cache allowing a total of 224MB of cache per controller. This gives you the ability to configure Synchronix to the optimum combination of cost and performance

SPECIFICATIONS
Disk Array
RAID Levels Multiple, simultaneous RAID Levels 0, 1, 3, 5, 10 & JBOD
Controller Supports dual initiation and provides array-based failover
Number of Disk Drives From 5 to 30 (with expansion chassis) per array; over 819GB in a single 70" rackmount cabinet
Usable Capacity Up to 273GB (with expansion chassis) per array; over 819GB per single 70" cabinet (three arrays)
Cache 16 - 224MB non-segmented adaptive cache (write-through or write-back) per controller, 1344MB within a single 70" cabinet. Includes the data consistency check which provides a three-phased commit of writes to ensure data integrity. LFU vs. LRU cache algorithm. Cache is fully user configurable via the GUI (*Assumes dual controllers on each of three arrays.)
Host Connectivity 40 MB/second Ultra Wide SCSI-3 differential
Hot-Swap Components Drives, RAID controllers, power supplies and fans
Array Management Software GUI RAID Manager; SCSI-based configuration, monitor and fault notification tools
Platform/OS Compatibility Hewlett-Packard HP-UX, IBM AIX, Microsoft Windows NT/Wolfpack, SUN: Unix and NT, SUN OS/Solaris, NCR and others
Database Compatibility Oracle, Sybase, Informix, and others
Disk Drives 2.1 GB 4.2 GB 9.1 GB
Form Factor 3.5 inch 3.5 inch 3.5 inch
Transfer Rate 20 MB/second 20 MB/second 20 MB/second
Rotational Speed 7200 RPM 7200 RPM 7200 RPM
Average Seek Time
(read/write)
8.0ms/9.0ms 8.0ms/9.0ms 8.0ms/9.0ms
Average Latency 4.17 ms 4.17 ms 4.17 ms
Cache Size per Drive 512 KB 512 KB 2048 KB
Interface UltraSCSI UltraSCSI UltraSCSI
Connector SCA SCA SCA
MTBF (power-on hours) 1,000,000 1,000,000 1,000,000
  • Solid State Disks (SSD) in various capacities up to 804MB are available as a media alternative in both non-volatile and volatile models and may replace a standard disk drive in one or more bays within a Synchronix array. SSDs are offered to complement the robust cache of the Synchronix SMS.

Options

Cache Upgrade Upgrade option from 16MB to 224 MB per controller
Expansion Chassis Provides 2nd and 3rd shelf of ten disk drives (30 drives total per array)
Redundant Controller Provides controller redundancy with host-transparent automatic controller failover
Global Hot-Spare Allows a single spare disk drive to provide automatic drive rebuild on all RAID groups in the array
UPS Interface for intelligent monitoring
ENVIRONMENTAL
Operating Temperature (Celsius) 0 to +40
Non-operating temperature (C) -20 to +65
Maximum rate of change (F): 5 per minute
Operating humidity (NC): 20% to 80%
Non-operating humidity 10% to 90%