PernixData: New storage statusquo is comming

Storage SME's knows for ages that storage design begins with performance. The storage performance is usually much more important then capacity. One IOPS  cost more money then one GB of storage. Flash disks, EFD's and SSD's changed storage industry already. But the magic and the future is in software. PernixData FVP (Flash Virtualization Platform) looks like very intelligent, fully redundant and reliable software cluster aware storage acceleration platform. It leverages any local flash devices to accelerate any back-end storage used for server virtualization. Right now only VMware vSphere is supported but solution is hypervisor agnostic and it is just a matter of time when it will be ported to another server virtualization platform like Hyper-V, Xen, or KVM.

PernixData setups absolutely new storage quality in virtualized datacenter. If you have issue with storage response time (latency) then look at PernixData FVP. But what impressed me is the future because I believe the platform can be improved significantly and new functionality will come soon. I can imagine data compression and deduplication, data encryption, vendor independent replication, clonning, snapshoting, etc.

So software defined storage virtualization just began.

Happy journey PernixData.

For more information look at
http://www.pernixdata.com/
http://www.pernixdata.com/SFD3/

Magic Quadrant for General-Purpose Disk Arrays

http://www.gartner.com/technology/reprints.do?id=1-1ENAPKJ&ct=130325&st=sg

Pretty nice overview and comparison among storage vendors. Because I have privilege to practically design, implement and work with many storage arrays I can't agree with IBM positioning and description. In the past I was also impressed about IBM storage products but reality is little bit different. I was troubleshooting several big issues with IBM mid-range storage array IBM V7000 (Storwize) and also with high-end IBM DS8700 (Shark).
  

Difference between SCSI-2 and SCSI-3 reservation

SCSI-3 reservations are persistent across SCSI bus resets and support multiple paths from a host to a disk. In contrast, only one host can use SCSI-2 reservations with one path. If the need arises to block access to a device because of data integrity concerns, only one host and one path remain active. The requirements for larger clusters, with multiple nodes reading and writing to storage in a controlled manner, make SCSI-2 reservations obsolete.

Info retrieve from:
http://sfdoccentral.symantec.com/sf/5.0/hpux/html/vcs_install/ch_vcs_install_iofence4.html

Reduced vCenter DB by deleting old events and tasks from vCenter database

In vCenter MS-SQL Database is storage procedure called cleanup_events_tasks_proc which deletes old data based on event and task retention settings. vCenter retention settings can be setup in vCenter Settings though vSphere Client or changed directly in database. Using vSphere Client  is recommended.

Following example is copied from: http://communities.vmware.com/thread/191227?start=0&tstart=0

c:> "C:\Program Files\Microsoft SQL Server\90\Tools\Binn\OSQL.EXE" -S \SQLEXP_VIM -E
1> use VIM_VCDB
2> go
1> update vpx_parameter set value='' where name='event.maxAge'
2> update vpx_parameter set value='' where name='task.maxAge'
3> update vpx_parameter set value='true' where name='event.maxAgeEnabled'
4> update vpx_parameter set value='true' where name='task.maxAgeEnabled'
5> go
(1 row affected)
(1 row affected)
(1 row affected)
(1 row affected)
1> exec cleanup_events_tasks_proc
2> go
1> dbcc shrinkdatabase ('VIM_VCDB')
2> go
DbId   FileId      CurrentSize MinimumSize UsedPages   EstimatedPages
------ ----------- ----------- ----------- ----------- --------------
      5           1       81080         280       78776          78776
      5           2         128         128         128            128

(2 rows affected)
DBCC execution completed. If DBCC printed error messages, contact your system
administrator.
1> quit