User Tools

Site Tools


brainstorms:research_computing_storage_infrastructure_2012

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revisionPrevious revision
Next revision
Previous revision
Last revisionBoth sides next revision
brainstorms:research_computing_storage_infrastructure_2012 [2012/05/25 08:59] – [Timeline] aorthbrainstorms:research_computing_storage_infrastructure_2012 [2012/07/24 06:39] aorth
Line 3: Line 3:
  
 ===== Current situation ===== ===== Current situation =====
 +  * HPC (June, 2011)
 +    * ~6TB of usable disk space, ~2.5TB in use right now
 +  * boran (database + VM server, January, 2012)
 +    * ~1.5TB usable disk space, ~20GB in use right now
 +
 +
  
 ===== Timeline ===== ===== Timeline =====
Line 9: Line 15:
     * Cassava genome project?     * Cassava genome project?
   * **May 24**: Alan and Isaac meet with NetApp storage representative ("GK" <gkumawat@techno-associates.co.ke>), facilitated by George Ogoti from ICT   * **May 24**: Alan and Isaac meet with NetApp storage representative ("GK" <gkumawat@techno-associates.co.ke>), facilitated by George Ogoti from ICT
-    * +    * Existing NetApp is expandable, there are various options we can explore 
 +    * GK is going to get us a quote for the following infrastructure 
 +      * RAID-DP (NetApp's version of RAID6, 2 disk failure) 
 +      * Site redundancy (storage syncs nightly via fiber to ICRAF) 
 +      * ILRI site will have two controllers for high availability 
 +      * ICRAF site will have one controller (less critical storage) 
 +      * Capacity 12TB or 24TB (with usable space roughly half of each figure) 
 +  * **June 13, 2012**: 
 +    * Got the quote back from GK at Techno Associates, two options: 
 +      * 12TB dual configurations: $40,000 
 +      * 24TB dual configurations: $48,000 
 +    * We need to talk to Ian Moore to see what he thinks 
 +    * It's possible we now use ICT's NetApp to provide 1-2TB for GIS server, then building some custom solution for the DMZ 
 +  * **June 14, 2012**: 
 +    * Brainstorming raw storage costs vs NetApp quote: 
 +    * <file>NetApp quote 1TB x 24 = 29000 USD 
 +NetApp quote 2TB x 24 = 39000 USD 
 + 
 +scan.co.uk 1TB Seagate 75 GBP x 24 = 1800 (~2800 USD) 
 +amazon.co.uk 1TB Seagate 65 GBP x 24 = 1560 (~2500 USD) 
 + 
 +scan.co.uk 2TB Seagate 70 GBP x 24 = 1680 (~2600 USD) 
 +amazon.co.uk 2TB Seagate 83 GBP x 24 = 1992 (~3100 USD) 
 + 
 +scan.co.uk 3TB Hitachi 130 GBP x 24 = 3120 (~4900 USD) 
 +amazon.co.uk 3TB Seagate 120 GBP x 24 = 2928 (~4600 USD)</file> 
 +  * **June 18, 2012**: 
 +    * Had a meeting with Ian Moore and Isaac Kahugu about storage 
 +      * He said he'd support us building our own, but gave us tips to talk to Tor at ICRAF (GIS, MySQL, Drobo), and to consider Dell Equalogic for storage 
 +      * Another point was that we could possibly buy storage from KENET (to sync off site), or maybe colocate a box there 
 +  * **July 10, 2012**: 
 +    * GK from Techno Associates called again with a new offer for a single-site, single-controller NetApp solution: 
 +      * He said he can give us 12TB raw for $9,000, or 24TB for $11,000 (one controller only, excludes pricing for replication licenses) 
 +  * **July 22, 2012**: 
 +    * Begin compiling report about current situation, options, and recommendations 
 +      * https://docs.google.com/document/d/123VL6l5xt1AspzqTaW2tpJ_XFUZGDIXUCsaFK2_EwUY/edit# 
 +  * **July 23, 2012**: 
 +    * George Ogoti provided us with an iSCSI target on their NetApp so we can test configuration and performance. 
 + 
 +===== Proposed NetApp architecture ===== 
 +Proposed architecture assuming we expand ICT's existing NetApp rack with extra controllers and storage. 
 + 
 +{{ :brainstorms:research_computing_storage_2012.png?nolink |}} 
 + 
 +**Key points**: 
 +  * Raw storage is sliced in several chunks and shared appropriately 
 +  * NetApp exports CIFS shares to corporate clients and servers (users authenticate with Active Directory credentials) 
 +  * NetApp exports iSCSI block devices to Linux servers in order to allow them to manage their own storage/access/users directly in the OS 
 + 
 +===== Alternatives ===== 
 +  * Build our own storage, ala Backblaze "pods": http://blog.backblaze.com/2011/07/20/petabytes-on-a-budget-v2-0revealing-more-secrets/ 
 +  * Using FreeBSD + ZFS? 
 +  * FreeNAS storage based on AMD Fusion APUs: http://the.only.ipnextgen.net/fnas/doku.php 
 + 
 +===== Links ===== 
 +  * Growing a ZFS pool (with good background on pools vs filesystems in ZFS): http://www.itsacon.net/?p=158
brainstorms/research_computing_storage_infrastructure_2012.txt · Last modified: 2012/07/24 06:57 by aorth