The Sun Oracle Database Machine is a million IOPS-capable platform. Why would a CIO consider it as a solution when a particular datacenter may not require so much I/O? Performance Architect in Oracle's Systems Technology Group Kevin Closson gives reasons and applies these to what really goes on in real world production datacenters, touching on chaos, IOPS, latency and Flash.

Using an example of 32 hypothetical Oracle Database instances deployed in the standard 1:1 (OS:DB) deployment model, Closson maintains these database instances would require 32 systems to maintain. Suppose these systems are older models consisting of 2 socket, multi-core processors that are three years old. He concludes that 32 databases could be deployed in a single Sun Oracle Database Machine and experience significant performance improvement. Consolidating the 32 instances into a database grid of 8 hosts (all Real Application Clusters-ready) would reduce a lot of chaos since there wouldn't be 32 pools of storage to manage.

"Disk space in the Sun Oracle Database Machine can simply be provisioned from one easily managed storage pool (Automatic Storage Management disk group)," he reasons. "That seems a lot simpler to me as does maintaining 24 fewer OS images."

He notes that the ability to run both RAC and non-RAC instances in the Sun Oracle Database Machine is another consolidation consideration worth value. The Database Machine makes switching a database from non-RAC to RAC a simple task because all the requirements are in place.

Another important point is the Database Machine allows for any of the databases to run on any of the hosts. Closson asserts that the hosts serve as a true grid of resources, and once the storage is central and shared, and all hosts interconnected, users essentially have a grid.

In terms of IOPS, consolidating databases into a Sun Oracle Database Machine allows IOPS-hungry applications to scale up to 8 nodes and 1 million IOPS with RAC. On the other hand, compute-light but IOPS-intensive databases can be hosted in a single database server in the Database Machine since there is a demonstrated 125,000 IOPS worth of bandwidth available to each host. "That’s over 15,000 per processor core," he explains. Although this may be more than is required, he surmises that database administrators are now relieved from having to specifically provision IOPS to various different database servers.

Closson also addresses those datacenters that may have situations where extreme IOPS are not necessary but very low latency I/O is. The fictional data set he uses is a 1 TB latency-sensitive database. His solution is to consolidate all sets of databases into a Sun Oracle Database Machine. He explains, "The lower-class 'citizens' can be stacked together inside one or a few database hosts and their I/O demand controlled with Exadata I/O Resource Management (IORM). The latency-sensitive 1 TB database (be it single instance or RAC), on the other hand, can operate entirely out of FLASH because the architecture of Sun Oracle Database Machine is such that the entire aggregate FLASH capacity is available to all databases in the grid. That is, databases don't have to be scaled with RAC to have access to all FLASH capacity. So, that latency-sensitive database can even grow to 5 TB and still be covered with FLASH and further, it can grow to become more processor-intensive as well and scale with RAC to multiple instances without procuring a new cluster."

All in all, Closson is most enthusiastic about the Flash storage-provisioning model offered in the Sun Oracle Database Machine because the Flash cache is dynamic. According to Closson, data flows through the cache where hot stuff stays and cold stuff leaves based on capacity. This solves the dilemma facing many database administrators, having to shuffle portions of databases in and out of Flash disks manually as hot spots move. "With the Database Machine, on the other hand, the problem sort of solves itself," he states.

More Information

Sun Oracle Database Machine: The Million Oracle Database IOPS Machine or Marketing Hype? Part II - Closson's blog entry

Sun Oracle Database Machine: The Million Oracle Database IOPS Machine or Marketing Hype? Part I - another Closson blog entry

Sun Oracle Database Machine with Sun FlashFire Technology

Technical Overview of the Exadata Product Family

Exadata Partner Program [...read more...]

fullsource

From the latest issue:

Recent Blog Entries as of March 22, 2010, 12:13 am
Oracle Works to Assuage the Anxieties of Its Customers SysAdmin News Bites Oracle Enterprise Manager Ops Center Released Suits and Their Countersuits A More Direct Channel Model Planned for Oracle-Sun	The Importance of Information Storage and Management Sun POJO Service Engine User's Guide Sun Storage 7000 2010.Q1 Quality and Features PSA Deploys GlassFish in the Manufacture of Peugeot, Citroën Vehicles Running VirtualBox VMs as Services in Windows
Blog RSS Feed