A couple of weeks back I was at Intel Cloud Day 2016 with the rest of the TFD team. We listened to a number of presentations from Intel Management team mostly about how the IT world was changing and how they planned to help lead the transition to the new cloud world.
The view from Intel is that any organization with 1200 to 1500 servers has enough scale to do a private cloud deployment that would be more economical than using public cloud services. Intel’s new goal is to facilitate (private) 10,000 clouds, being deployed across the world.
In order to facilitate the next 10,000, Intel is working hard to introduce a number of new technologies and programs that they feel can make it happen. One that was discussed at the show was the new OpenStack scheduler based on Google’s open sourced, Kubernetes technologies which provides container management for Google’s own infrastructure but now supports the OpenStack framework.
Another way Intel is helping is by building a new 1000 (500 now) server cloud test lab in San Antonio, TX. Of course the servers will be use the latest Xeon chips from Intel (see below for more info on the latest chips). The other enabling technology discussed a lot at the show was software defined infrastructure (SDI) which applies across the data center, networking and storage.
According to Intel, security isn’t the number 1 concern holding back cloud deployments anymore. Nowadays it’s more the lack of skills that’s governing how quickly the enterprise moves to the cloud.
At the event, Intel talked about a couple of verticals that seemed to be ahead of the pack in adopting cloud services, namely, education and healthcare. They also spent a lot of time talking about the new technologies they were introducing today.
Intel Xeon E5 V4 processors
Intel showed off their new E5 V4 Xeon processing chips with 22 cores, DDR4 memory, virtualization improvements, interrupt processing improvements, ABX transactional synchronization technologies, and more, all within the same power envelope as previous generation processors. One interesting feature was the Intel’s Resource Director which allows one to apply QoS to L3 caching, so that noisy neighbors don’t boot out higher priority processor workloads out of L3 cache.
Network Builders Program
Intel said that the Network Builders has 100s of partners currently and the builders program has helped in creating 40 new network virtualization functionality deployments with 100s more in the pipeline. And the Network Builders Program has 5 of the top 10 telecoms world wide in the program. Essentially, the Network Builder Program supplies technology, reference architectures, standards development, GoTo Market services and matchmaking capabilities to help further adopt and deploy network function virtualization technologies.
Storage and Cloud Builders Program
Announced at the show, the Storage Builders Program already has a number of partners. Intel highlighted a few, UBS-Coho Data, University of Michigan-YottaByte, and QCT-VMware but they had a chart with over 70 current partner logos on it, which included every major and most startups in the storage industry. The intent of the Storage Builders Program is to help address industry wide issues, by exploiting collaborative efforts to drive enterprise grade performance, scale and reliability/availability into Software Defined Storage and other disruptive storage architectures, and to help validate and accelerate the adoption of these new technologies.
The Cloud Builders Program was probably the least described of the three, but still they showed a chart with over 20 logos on it. The hope is that these partners can help deploy the next 10,000 clouds using Intel technologies.
New SSDs and PCIe NVMe storage
The other big news of the day, from the storage perspective is Intel’s new SSDs, two new data center planar NAND SSDs, the D3700 and D3600, and two new 3D NAND drives, the P3520 and P3320 SSDs, all of which connect via PCIe and use NVMe as their storage protocol.
The new D3600 is a higher capacity (1 and 2TB) SSD with lower endurance (3 DWPD) while the D3700 is higher performing, lower capacity (800GB and 1.6TB) with higher endurance (10 DWPD). Intel says the new NVMe SSDs support almost 4X more 4K random read IOPS and roughly 2X more read and write bandwidth then similar SAS attached SSDs.
The new P3520 and P3320 NVMe devices utilize new, 3D MLC NAND technology and also have high performance, although Intel didn’t provide any specifics here. And there weren’t any DWPD specs provided either. But a follow on question said they were 1 and 0.3 DWPD devices.
Behind the news is a very public push to support Non-Volatile Memories. Microsoft just announced support for this technology as did some Bios vendors. The view in the TFD team was that this is all in preparation for 3D Point to emerge as a new memory architecture. Check out the Storage Roundtable at Intel Cloud Day 2016 to hear more on this topic.
As for NVMe, more devices are coming out every day. I asked Intel if they were hot pluggable and they said yes. But during the following weeks, when I was talking with another storage vendor, they said that the current NVMe devices (maybe not Intel’s new NVMe drives) were not surprise hot pluggable. That is someone couldn’t go up to a device and unplug it without causing system problems. However, if the system knew beforehand (pre-approved [?] hot pluggable) that a device was going to be unplugged, it could take action to disable the interfaces on the device (or bus connection) so that it’s removal wouldn’t present a system problem. Seems there’s still some work in making NVMe drives truly hot pluggable.
This week I am off to Austin to hear what Micron is up to with their 3D NAND, 3D Xpoint memories and their own NVMe SSDs.