<   Previous Post  Heartbleed and IBM
User Group Mania  Next Post:   >

Comentários (20)

1 jdkruse comentou às Link permanente

Barry, <div>&nbsp;</div> Exciting news. Is there a v7.3 draft redbook floating around for V7000 and SVC? <div>&nbsp;</div> jk

2 JonTate comentou às Link permanente

Hi Jordan, we are currently writing the IBM Redbook on the DH8 as I type, and that will highlight the salient features of the DH8 and 7.3. We also have it in plan to cover the Storwize V7000 Gen2 in the next month or so. I expect us to have a draft of the DH8 available in the next month or so all being well.

3 talor comentou às Link permanente

Hi Barry, <div>&nbsp;</div> The hardware improvements sound really good, is there anything on when 16G FC will be available on SVC/V7000? <div>&nbsp;</div> Also with the software updates, this is great as at removes the need for SVC tools to balance extents across all mdisks in a pool, is there anything in 7.3 making an easy way to reclaim space on thin provisioned volumes rather than doing vdisk mirror/unmirror? <div>&nbsp;</div> Finally Jon, Tony Pearson's blog mentioned on V7000 unified the need for SFU being removed in 1.5 code, is the V7000 Unified also covered in the updated redbook?

4 4R18_Nicholas_Savva comentou às Link permanente

Are there any plans to integrate a scheduling utility for FlashCopy? this is being done using external workstations/scripts etc.

5 orbist comentou às Link permanente

talor, we have issued a "statement of direction" for 16Gbit FC connectivity, but further details would need an NDA etc <br /> I'll chase Jon re the Unified redbook.

6 orbist comentou às Link permanente

Nocholas, IBM has the FlashCopy Manager software that can intergrate into applications, and provide much value added benefits for triggering, scheduling and controlling the use of FlashCopy. Therese type of functions need to be closer to the applications to be able to trigger a consistent copy, of which the storage has no real idea. So at present there are no plans to add scheduling to the storage products themselves.

7 rankink comentou às Link permanente

When can we expect more granular monitoring out of the V7000? Say, per-volume statistics?

8 Tomasko comentou às Link permanente

Barry, <div>&nbsp;</div> - what number RTC we can expect I mean upper 8KB iops limit for one io group ? actually according http://www-03.ibm.com/support/techdocs/atsmastr.nsf/WebIndex/PRS5183 this is 45000 iops 8KB dual cpu <br /> - curently in CG8 is not possible to use more than 4 cpu for system cores (8 for io group) I have suspicions that iops, throughput is limited by system cores but when we tested SVC + flash system 820 I haven't seen greater system cpu utilization than 80 % and we got only 2,7 GB/s througput (went down from 3.2 GB/s direct flash without SVC) <div>&nbsp;</div> thanks

9 daboule comentou às Link permanente

Hi, <div>&nbsp;</div> do you know when scsi unmap will be implemented on next v7000 engine ? <div>&nbsp;</div> there is a feature request here that you may already know : <div>&nbsp;</div> http://www.ibm.com/developerworks/rfe/execute?use_case=viewRfe&amp;CR_ID=26193

10 JonTate comentou às Link permanente

@talor - we do not include any updates to the Storwize V7000 Unified in the 2145-DH8 book. Which is coming on rather nicely. I'll knock something up for this blog once I publish the draft.

11 Tony888 comentou às Link permanente

I have a question on V7000 v2 <div>&nbsp;</div> V7000 V2 can scale up to 504 drives per IO group and 1056 drives in a four-way clustered. <div>&nbsp;</div> How may drives can scale up to in a two-way clustered? Is it 1008 drives ? <div>&nbsp;</div> How many drives can scale up to in a three-way clustered? Does it can reach 1056?

12 orbist comentou às Link permanente

Tony, you are correct : <div>&nbsp;</div> Each IOGroup (control enclosure) can have at most 504 drives attached to it <br /> Each cluster (up to 4 control enclosures) can have at most 1056 drives attached to it. <div>&nbsp;</div> So : <br /> 1 IOG = 504 <br /> 2 IOG = 1008 <br /> 3 IOG = 1032 <br /> 4 IOG = 1056 <div>&nbsp;</div> If you are going up to 3 or 4 IOG, then suggest even spread of the expansions across IOG, and when we support the logically maximum 2016 drives in the future, you can expand each IOG evenly.

13 OlegKorol comentou às Link permanente

@rankink <br /> for more granular monitoring - you can use tivoli tpc <br /> or free stor2rrd - www.stor2rrd.com

14 scott.fuhrman comentou às Link permanente

Now that 7.3 has been released, when will we see the new STAT tool?

15 ivan.klimov comentou às Link permanente

Is it possible to share slides from Edge sessions sTU01/02 and sDS01/02? <br /> Thanks!