background image

 

 

©2008 The Tod Point Group

 

www.TheTodPointGroup.com 

 

 

Page 

2

Architectures 

IBM  

Named the IBM TS7700 Virtualization Engine, this new system is a significant change 
from the previous architecture. IBM calls this new architecture its “Virtual Tape Grid” 
and, while it continues to be based on outboard subsystems, the way nodes operate 
and are managed is totally changed. At its heart, this grid is a move away from a 
monolithic approach to a more distributed one. Unlike IBM’s old peer-to-peer 
architecture, the individual TS7700 nodes are managed as an integrated whole. 
Moreover, each node adds capacity to the local grid and sports new gigabit IP links for 
inter-node traffic. 
 
Another important architectural difference is that IBM now virtualizes the location and 
number of virtual volumes. This permits access and recovery from any node and was a 
key design goal. The idea is to completely divorce the user from having to know where 
tape data actually resides and how many copies there are. Instead, users can specify 
recovery policies that will ultimately dictate these. As we discuss below, this added 
virtualization dimension is not only a big step in supporting business continuity with 
advanced policy management; it is also a necessary step in content-based access for 
tape data. 

 

User interest in the new gigabit IP links has been strong as they look for simpler and 
less expensive ways to replicate tape data. IBM led the market with synchronous 
replication and has always had asynchronous replication. 
 
Although IBM’s tape grid currently supports three nodes, it has been designed to 
support at least eight nodes as well as to support future enhancements in cache 
capacity, performance, content-based access and data de-duplication. It also leverages 
IBM’s vast array of technologies for performance, encryption and futures. 
 
Playing catch-up, IBM has finally added the capability to export a copy of the logical 
volumes stored to the node for disaster recovery purposes. Also, IBM’s tape grid is not 
compatible with its previous architecture, but a wealth of migration tools are available. 
 

Sun 

 
Sun continues to employ a hybrid approach where most of the virtual tape logic sits on 
mainframe-based software while tape I/O and replication services are handled by a 
relatively dumb external hardware subsystem based on Sun’s SVA array which is no 
longer being actively marketed.  
 
In addition, availability features are limited to two-node clusters with no more than four 
nodes total (one local cluster and one remote cluster) often referred to as a “quadplex”. 
Worse yet, only one set of channels must handle all I/O (host, tape drives, and remote 
communication) making it difficult to balance changing workloads. In addition, without 
gigabit IP links, expensive channel extenders or routers must be used to reach remote 
sites. 
 

Отзывы: