I’d like to point everyone over to a great review of the Calxeda-powered Boston Viridis box by Anandtech that just went live, here. First of all, big thanks to Johan De Gelas over at Anandtech and Wannes De Smet at SizingServers for doing a top notch job pulling together an in-depth review of our gear as well as the team at Boston Limited for taking care of the hardware. Since we launched the ECX-1000 we’ve been beating the streets to get real results and metrics out into customers’ hands and show that the technology delivers as promised. With quotes like “Calxeda really did it”, “nothing short of remarkable” and “revolutionary technology”, we’re all excited to see these results posted on a site like Anandtech.
Anandtech Reviews the Calxeda ECX-1000: “Calxeda’s ECX-1000 server node is revolutionary technology”
As reported in various outlets yesterday, Intel has released their S1200 line of Atom SOC’s targeting the microserver market with the tagline: “Intel Delivers the World’s First 6-Watt Server-Class Processor”. The first notable point here is that they had to use 6 Watts, because 5 was already taken. The second notable point is their definition of “Server-Class”. Looking at the list of features on the Atom S1200, there are key “Server-Class” features missing:
- Networking: Intel’s SOC requires you to add hardware for networking
- Storage: Once again, there is no SATA connectivity included on the Intel SOC, so you must add hardware for that
- Management: Even microservers need remote manageability features, so again with Intel you need to tack that on to the power and price budgets.
Based on what Intel disclosed today, here’s a snapshot of Calxeda EnergyCore 1000 vs. Intel’s new S1200 chip:
|Cache (MB)||4 Shared||2 x .5 MB|
|PCI-E||16 lanes||8 lanes|
|Fabric Switch||80 Gb||NA|
|Address Size||32 bits||64 bits|
|Memory Size||4 GB||8 GB|
So, while the Centerton announcement indicates that Intel takes “microservers” seriously after all, it falls short of the ARM competition. It DOES have 64-bits and Intel ISA compatibility, however. Most workloads targeting ARM are interpreted code (PHP, LAMP, Java, etc), so this is not as big a deal as some would have you believe!Intel did not specify the additional chips required to deliver a real “Server Class” solution like Calxeda’s, but our analysis indicates this could add 10 additional watts PLUS the cost. That would imply the real comparison is between ECX and S1200 is ~3.8 vs ~16 watts. So roughly 3-4 times more power for Intel’s new S1200, again, comparing 2 cores to 4. Internal Calxeda benchmarks indicate that Calxeda’s four cores and larger cache delivery 50% more performance compared to the 2 hyper-threaded Atom cores. This translates to a Calxeda advantage of 4.5 to 6 times better performance per watt, depending on the nature of the application.
Calxeda’s approach to driving power optimization in the datacenter goes well beyond the processor. We focus on enabling our partners to achieve rack level power efficiency based on our technology. Last week, Boston Limited announced their 2U Viridis platform with 24 Calxeda EnergyCore(TM) server nodes, 96GB of memory, and 6TB storage is measuring 130W “at the wall”. This equates to just 5.5W of power per server inclusive of memory, disk and chassis-level overhead. At a fraction of the power of a traditional x86 server node, the Viridis server cluster based on Calxeda EnergyCore will allow datacenter operators to experience an order of magnitude improvement in efficiency. [Read more...]
Back in June, Calxeda published web-serving benchmarks that claimed a significant advantage in performance per watt over x86-based servers. Using ApacheBench, a single 5.26 watt Calxeda EnergyCore server delivered 5500 transactions per second, compared to a 102 watt (TDP) Intel E3-1240 that saturated the network at 6950 TPS. About 2 months later, Intel spoke with Timothy Pricket Morgan at The Register to provide their response.
You have to hand it to Intel; they make really fast processors, which are appropriate when maximum compute performance is needed. But Intel’s argument is missing the point, the very reason why Extremely Efficient Servers are a promising trend: by right-sizing the compute, memory, and networking infrastructure to meet real workload requirements, one can save a great deal of money and power. Intel’s response is classic PC-Server era thinking: use a faster CPU, and then feed it like a force-fed goose being prepped for foie gras. In this case they added a 10G ethernet port to try to close the gap. But if 5000 transactions per second is all your website needs, or you use load balancing to handle the peak loads above normal usage, Calxeda is dramatically more efficient. That is the point.
It is a bit surprising Intel went to these lengths when Intel’s own math shows that Calxeda maintains a 4-5X performance/watt advantage versus the solution most websites would use. Apparently not satisfied, Intel then upped the ante and added an expensive 10 Gb network infrastructure to keep their uber-fast processor busy. With this configuration, Calxeda is still some 30% more efficient than the significantly more expensive* 10Gb Ivybridge solution. But small-medium web sites rarely use or need a 10Gb ethernet port; a 1Gb interface is usually sufficient for typical demand. Moreover, Intel’s proposed alternative would require two 10Gb top of rack switch (TORS) ports in addition to the 3 NICs (2 for data, 1 for management). Those TORS ports alone could add 10-15 watts per server for the 10Gb solution that were not included in Intel’s math. But hey, it won the benchmark (well, almost)!
Calxeda is focused on providing energy-efficient solutions for real-world problems and we believe that bigger and faster is not always better. Leaner and cleaner can be less expensive and far less power hungry, lowering costs for real-world workloads which can be highly variable. Which is more representative of your real real-world environment? You be the judge.
* Based on comparing the servers w/o disks to isolate the server-power, and adding 1 watt to each 5.26 watt Calxeda node to estimate wall power, assuming a modest 24 nodes in a chassis share the power supply and fans. Note that each Intel server equipped as Intel suggests would require a PCI extension with 10 Gb NICs, and switch ports; 2 for data and 1 for management. These are costly additions ($700 per 2 ports, plus the required 10Gb TORS ports) to the IvyBridge server, and of course consume even more power. We are still optimizing our platform and Calxeda will publish a slew of benchmarks and wall-power measurements in the coming weeks.
It’s the middle of June, which means we’re smack in the middle of tradeshow and conference season for the IT industry. We were at Computex in Taipei two weeks ago, and this week we’re participating in International Supercomputing in Hamburg, and GigaOM’s Structure conference in San Francisco. In fact, our CEO, Barry Evans, is on a panel to discuss fabric technologies and their role in the evolution of datacenters. Should be a good one!
In spite of the hectic season, it hasn’t stopped us from moving forward with what everyone is really waiting for: benchmarks! Well, I’m happy to be able to share some preliminary results of both performance and power consumption for those of you looking for more efficient web servers.