for OMD reference :)
We can see the intended shape of Violin as it emerges fresh and shining from its refit, but we've yet to hear it played. What will be the next move of Violin's brand new CEO, Kevin DeNuccio? DeNuccio has been restringing Violin, cutting and restructuring the organisation's engineering and go-to-market, with more emphasis on …
for OMD reference :)
Indeed, although I was rather disappointed to find this was not an article discussing the relative merits of the electro-tech employed by the seminal electro-popsters.
Still, I also like the reference.
Me too, you could have at lease put a synth reference half way through to keep my interest... PANTS!
...should be the first one on the fix-list, I think: with a promised ~3GB/s max throughput per an all-flash array (as I was told during a recent demo) priced at couple hundred Ks I think the use of "high-performance" is questionable... high-end? In terms of technology or price it is, for sure. Scale-out? Perhaps, with the optional built-in 2-node file sharing cluster definitely, fine. High-performance? Not with the reported 3GB/s max for that price, sorry, I don't think so. Not sure what happens to all those crazy-fast flash modules, all plugged directly into memory slots, but to be able to really compete on this crowded market I think they need a lot better throughput OR lower its price significantly - I've seen Tegile, Fusion-io etc offerings deliver the same or better throughput and they both cost less than Violin's 64TB 6424 w/ Windows Storage Server 2012 R2 cluster. (FWIW I realize that FIO's ioControl array is block-only but at the same time even their uber-SPX bundle costs probably around half of a 6464 and Violin's optional offering of two quad-core built-in blades running WSS 2012 in scale-out file sharing cluster mode utilizing the array as CSV does not really warrant another 100k+ added to the price tag.) For that kind of money one can even go to IBM and get their shiny new (first non-ex-TMS) FlashSystem 840, scaling up to 8GB/s - now that sounds more like "high-end high-performance" to me...
fscked by SHA-1 collision? Not so fast, says Linus Torvalds