This week Network World laid out some details of the work the IEEE group, the Ethernet Alliance, is doing with respect to new data rates. As mentioned in this blog post, while there are 5 shipping speeds of Ethernet (100 Mbps, 1 Gbps, 10 Gbps, 40 Gbps, and 100 Gbps), there are 5 new speeds that are currently being worked on (2.5 Gbps, 5 Gbps, 25 Gbps, 50 Gbps, and 400 Gbps). The last time Ethernet got this sexy was when promiscuous mode was introduced.
Some of the drivers for these new speeds are adoption rates of the older speeds. As detailed in the July 2014 IEEE Call for Interest , while the initial adoption for 10G, 40G, and 100G was in 2004, 2012, and 2015 (anticipated) respectively, because these speeds are turning out to be cost prohibitive, the transition to higher speeds has been slower than previously forecasted. For example, the 1G -> 10G transition has repeatedly moved out (from 2012 to 2014 to 2016 now). This creates a window where new technology can provide the higher port speed at lower cost. So, as an example, the SFP+ technology can be leveraged in 25 Gbps as a single lane and 50 Gbps as two lanes.
The 2.5 and 5 Gbps speeds (known as MGBASE-T) address the growing demands of BYOD in campus networks. Many of the newer APs nowadays ship with 802.11ac. This Wifi standard will have a second wave in 2015 whereby the uplinks (or backhauls) between the APs and the access switches will be multi-gigabit rates. The key requirement here is to be able to reuse the existing cabling infrastructure. So Cat 5e and Cat 6 would still be supported over the usual 100 meters and there would be no need to rip and replace cables.
Ethernet has come a long way since the days of the 2.94 Mbps flavor that Bob Metcalfe had invented. There is very little in common between the types of Ethernet standards we have today from the IEEE and the original specification. One thing that is common, however, is the ability to evolve according to market needs, from single-pair vehicular Ethernet to four-pair PoE and in between. More on this in another post.
Recently news broke out about Windows Server introducing support for Docker. This is significant because the ultra hot company had previously only been supported on Linux (and Azure). One of the major complaints about it was the lack of flexibility when it came to host operating system support. With this news Microsoft also announces that it will be contributing to Docker’s open source APIs. What a remarkable change from a company that epitomized closed systems.
I’m excited to announce today that Microsoft is partnering with Docker, Inc to enable great container-based development experiences on Linux, Windows Server and Microsoft Azure. Docker is an open platform that enables developers and administrators to build, ship, and run distributed applications. Consisting of Docker Engine, a lightweight runtime and packaging tool, and Docker Hub, a cloud service for sharing applications and automating workflows, Docker enables apps to be quickly assembled from components and eliminates the friction between development, QA, and production environments. Earlier this year, Microsoft released support for Docker containers with Linux on Azure. This support integrates with the Azure VM agent extensibility model and Azure command-line tools, and makes it easy to deploy the latest and greatest Docker Engine in Azure VMs and then deploy Docker based images within them. – Scott Guthrie, executive vice president of the Microsoft Cloud and Enterprise group.
Ivan Pepeljnak’s makes an important point in his webinar on Cloud Computing Networking: as a customer, understand the QoS and SLA Guarantees that your public cloud provider offers. Whatever Tenant A does should not impact the performance of Tenant B. At a very minimum, there should be some guarantees on bandwidth, IO operations, and CPU cycles for every tenant. You don’t want to have the noisy neighbor who hogs up resources that leaves you no choice but to reboot your VM with the hope of getting reassigned to a physical server with less load. An AWS Small Instance is an example of an environment where you might encounter this scenario.
Recently I stumbled upon the blog of David Gee in the UK. He covered the Cavium acquisition of Xpliant as well as Broadcom’s announcement of the StrataXGS Tomahawk chipset less than two months later. The remarkable thing about both chipsets is that they are both capable of 3.2 Tbps and feature programmability, something which the Trident II (a 1.28 Tbps chipset) didn’t have. The Trident II is used on Cisco’s Nexus 9000, Juniper’s QFX5100, and HP’s 5930, to name a few switches. There had been great anticipation for the Trident II because it contains support for VXLAN, which the Trident did not. However, the most recent tunnel encapsulation protocol, Generic Network Virtualization Encapsulation (GENEVE), isn’t supported on Trident II. Well, with Tomahawk, as well as Xpliant, because of their programmable nature, they should, in theory.
Broadcom’s press announcement page contains an impressive array of quotes from vendors such as Brocade, Big Switch, Cumulus, HP, Juniper, Pica8, and VMware, to name a few. It remains to be seen what vendors will implement Xpliant.
Earlier this week, news broke out on SDNCentral about a new startup called SocketPlane that integrates Docker containers with Open vSwitch (OVS). Docker is one of the hottest areas in enterprise tech these days. At the OpenStack SV event last month, Mirantis CEO Adrian Ionel, said that Docker had had 20 million downloads in the past four months mainly due to its ease of use and its benefits to developers. He showed a screenshot of Google Trends with ‘Docker’ compared against ‘Virtualization’. That picture is recreated below.
One of the co-founders of SocketPlane is Brent Salisbury, who has a network engineering background in academia before joining Red Hat earlier this year. In recent years he got more involved in the Open Daylight (ODL) project and is arguably the most well known network engineer-turned-coder. His blog has a wealth of information on hands on guides for installing and integrating OVS, OpenStack, and ODL, which I’ve referred to frequently. Two other prominent contributors to ODL, Madhu Venugopal and Dave Tucker, are the other co-founders of SocketPlane.
I had listened to a Class C Block podcast on ODL in November 2013, in which Venugopal and Salisbury spoke at length of their involvement with the project. Definitely worth a listen if you have the time.
Recently, Big Switch Networks earned bragging rights as the first networking vendor to attain the OpenStack Compatible certification. To achieve this status, the requirements are different for hardware and software products. Big Switch demonstrated compatibility with both Nova and Neutron networking environments. There are more details on the Big Switch and Mirantis sites. Big Switch differentiates between the two environments as:
- In a Neutron implementation, the Big Cloud Fabric leverages the BSN ML2 Driver, enabling automation and orchestration of its bare metal, SDN-based Big Cloud Fabric with the OpenStack controller.
- In a Nova implementation, Big Cloud Fabric has optimized configurations and performance enhancements that let it serve as a multi-path leaf/spine CLOS service 4k VLANs to every edge port. Unlike traditional spanning-tree based switching designs, full cross-section bandwidth can be acheived while delivering 4k vlans to every edge port with no performance penalty.
The question I have is that while obviously somebody has to be first, why aren’t there more products and vendors listed? Specifically, how soon will it be before we see HP, the leading contributor to OpenStack on that list?
TL; DR – They just follow the laws of Physics.
Moore’s Law states that the number of transistors per integrated circuit will double every two years. In a recent interview with Marketplace, Intel CEO Brian Krzanich, who is the sixth CEO of the company, expressed hope that Moore’s Law would remain alive on the watch of the next couple of CEOs. Currently, Intel can achieve 14 nm manufacturing processes. Krzanich said with the current technology, they can keep Moore’s Law alive for another 6-10 years.
Recently, I’ve been reading up on the science behind cricket bat manufacturing. Cricket has increasingly become a batsman’s game, highlighted not only by higher scores, but by bigger hits (more fours and sixes being hit). There are a few prevalent theories about bat manufacturing techniques improving and bats becoming heavier. A heavier bat can result in the ball being hit farther. However, as described by bat-maker Chris King, “the material and design have been pushed to their limits. Like Formula One cars, they operate at the outer edges of what’s possible.” Instead, the psychology of bat-owners is where the innovation lies. Bats, which have always been made of willow, aren’t necessarily getting heavier, but rather bigger and lighter, to give a batsman the impression that it’s heavier. Variations in the balance and the weight distribution make a huge difference. And that is what fives the batsman the sense of security and confidence to go for bigger hits. As King says, “What we’re up against is the belief that a big bat is more powerful than a bat of the same weight that’s smaller, which it isn’t. That’s against the laws of physics.”