Thoughts on a Thursday

We’ve been doing the startup thing for a hair under 13 years now. Most of the time we’ve been self funded, and recently we took a small investment in a friends and family round (angel.co link here).

What occurs to me, after we soft announced our 100GbE results via a Mellanox PR today, is that we’ve been building the types of high performance platforms that enable end users to do bigger and better things for the whole time. Our focus has always been on performance, and doing everything possible to make sure end users can make effective use of it.

Performance isn’t as simple as strapping in an SSD or NVMe and calling it a day. Good performance comes from good design, good implementation. You can’t start out with a poor design, add software and instant big data or hyperconverged system (though some folks would really like you to believe this). Performance is an absolute necessity now and going forward, and crappy designs just won’t play.

In the same vein, as much as some folks might claim, this is simply not a race car.

Sort of like BASF, we don’t make the products our customers make, we simply make it possible for them to deliver them faster, better, more accurately, by providing overwhelming computational, storage, and networking firepower. Providing this capability is what we live for. This is our reason for existence.

And its pretty cool.

We aren’t a network company, but our siRouter may well be one of the fastest SDN devices used in market.

Our storage products are being used to accelerate genomics computing at many sites.

Our hyperconverged cloud products are the basis of public clouds.

We like people to think about what they could do if we could reduce the impact of the IO/computational/data motion wait times. If we could enable more efficient computing and storage.

I am very positive about the direction we are going, and I gotta say, its nice to see industry and users align along the directions we’ve been talking about for the last decade plus.

Data motion is hard, so exploit data locality as much as you can, and when you can’t, have massive data pipes and data flows to be able to move a metric ton of it per unit time.

Now off to class before I am late …

Viewed 26949 times by 1239 viewers

Interesting conversation with a customer about our siRouter

They are turning their SDN concept into one of the most incredible technologies around, a tremendous competitive advantage for them over others in their space. I had been under the impression that they were running everything on their (quite awesome) 10/40GbE switches. These are SDN capable switches from a very well funded SDN switch startup.

Turns out, their SDN stack is actually running on siRouter. They are doing some very cool bits on the software stack side, and getting about 2 microseconds port to port. On Gigabit ethernet.

I was blown away.

Viewed 26726 times by 1118 viewers

Our 100GbE flash storage appliance benchmarks discussed

See the PR bit here (http://www.hpcwire.com/off-the-wire/new-mellanox-performance-benchmarks-released/ for the link impaired)

This is a Unison Ceph appliance ( http://scalableinformatics.com/unison ) and they are available and shipping now. Please reach out to us if you’d like to discuss.

And yes, this is the world’s first 100GbE storage appliance, or storage server SAN device if you prefer. Easily one of the fastest systems in market.

[Update] Forgot to mention, this is a set of units bought by a customer, and at their site. Not allowed to mention their name, though they are very well known.

Viewed 27229 times by 1133 viewers

Day job is hiring

Business development/sales role for now. See here (url: https://scalableinformatics.com/bus-dev in case you don’t see the link) for more details.

Prefer New York, Chicago, Boston, or nearby. No relocation.

Viewed 44503 times by 1661 viewers

SIOS v2.0 running pxe booted

Our SIOS (Linux based OS, usually based upon Debian) has just been updated for jessie (Debian 8). This was necessary to support rkt, docker, etc. in addition to our other bits. Its been cooking in the background for a while, for, as you might have noticed from my posting frequency, I’ve been busy.

But we are up, and running. Base distro version here:

root@usn-ramboot:~# df -h
Filesystem      Size  Used Avail Use% Mounted on
tmpfs           8.0G  2.7G  5.4G  34% /
udev             10M     0   10M   0% /dev
tmpfs           3.2G  9.2M  3.2G   1% /run
tmpfs           7.9G     0  7.9G   0% /dev/shm
tmpfs           1.0M     0  1.0M   0% /run/lock
tmpfs           7.9G     0  7.9G   0% /sys/fs/cgroup
tmpfs           1.0M     0  1.0M   0% /data
tmpfs           4.0G     0  4.0G   0% /var/tmp
tmpfs            64K  4.0K   60K   7% /var/lib/nfs
tmpfs           4.0G     0  4.0G   0% /tmp
tmpfs           1.6G     0  1.6G   0% /run/user/0
root@usn-ramboot:~# uname -r
3.16.0-4-amd64
root@usn-ramboot:~# cat /etc/debian_version 
8.1

Baseline version always first, and we’ll work from there.

Viewed 43676 times by 1339 viewers

M&A: Avago grabbed Broadcom, Intel grabs Altera

Avago continues its acquisition spree. Broadcom (network chipsets and NPUs, CPUs, etc.). This is looking like a more integrated semiconductor IP play here. They grabbed LSI, and shed the non-chippery bits. They grabbed PLX. And Emulex.

As they say, curiouser and curiouser. This makes perfect sense to me, and given the other acquisition announced today, I am going to bet they will be talking (at least) to Xilinx.

And then there’s Intel. Who just announced a purchase of Altera.

Viewed 59146 times by 1589 viewers

M&A [RUMOR]: Cisco grabs Nutanix

[update] TL;DR this appears to be rumor/speculation. One would think that such an acquisition would be prominent on Nutanix’s web site.

Its April fools, in May.

/sigh


Huge in the hyperconverged space (which, not so curiously, is where the day job is), and its setting up the battle lines between the major software/hardware players.

Cisco was already number 5 hardware vendor, and was bragging about “beating the white boxes“.

The last may be more wishful thinking than reality. I am not seeing much in the way of large scale infrastructure players deploying UCS. Rather its folks whom want a simpler converged infrastructure between switching and computing, when performance and capability aren’t the primary motivations, but purely management and single support phone number.

That is, Cisco has become, to some degree, the IBM of old.

As I remember, this did not end well for IBM. It took a while, but competitors found holes in the armor, and exploited them relentlessly. Here, Cisco doesn’t quite have the lock-in IBM had. So its declaration that it has killed off non-tier-1 nameplate systems is, well, at best, highly premature.

This said, Nutanix was working closely with Dell. Dell has a switch (Force10) and related capability. Cisco has UCS.

Curiouser and curiouser. As the article correctly notes, some folks will be badly injured by this, but I am not convinced their analysis is as thorough as they needed it to be. There is a secular change going on in the market, with the rise of hyperconverged systems, and their related bretheren, server SAN. Wikibon does, as usual, a bang up job on the analysis. Other sites provide good commentary on the analysis.

Basically, filers and alike, the architecture of the preceding decade, is dying. Albeit slowly, it is not in growth mode.

What this means is, given the growth in the hyperconverged space at 162% CAGR (!!!!!) is that the sea change in storage/computing is being driven by the hyperconverged appliances, and that organizations are in the long tail of providing support to legacy architectures for the existing SAN/NAS domain.

This is an existential challenge, not merely injurious, for those without a sane server SAN strategy. Nutanix started out as a hardware/software fused appliance, and migrated to become a software stack atop Dell hardware. Now UCS. But one thing we’ve noticed for years in our own tightly coupled aka hyperconverged systems … you can’t build these systems on crappy foundations. You need … as in you absolutely must have … performance for these systems to provide a meaningful impact. Which keeps much of the “lets build to the least possible acquisition/sales cost” crowd at bay, as they would require so many units to achieve the performance/capacity that we can achieve in far fewer, that it becomes cost ineffective to give serious consideration to these inefficient platforms.

Talk about validation of our ideas.

I do expect more M&A. Lots more. There are some folks with serious differentiation (gulp) and folks with software stacks.

Viewed 98763 times by 2204 viewers