Fergus Perry, Citi, Collateral Optimization, here.
Recent industry papers have estimated the additional collateral burden demanded by the Dodd-Frank and European Market Infrastructure Regulation (EMIR) legislation to be over $2 trillion globally.1 A major component of this increase results from the progressive move to central clearing of OTC derivatives and the corresponding requirement for all market participants to post initial margin to the central counterparties (CCPs) as part of the protection required by the CCPs against broker default.
Ari Balogh, Google Developers Blog, Google Compute Engine is now Generally Available with expanded OS support, transparent maintenance, and lower prices, here. Curious how much of the floating point instruction execution capacity is preserved when you execute your optimized code through a VM. A couple of the benchmarks that I came across on google compile some code -O2 and then present the Non-VM versus the VM performance. Not necessarily the most illuminating test. Other benchmarks simply show the relative performance of various VMs doing executing some floating point code. Again does not really give a handle on the problem.
New 16-core instances
Developers have asked for instances with even greater computational power and memory for applications that range from silicon simulation to running high-scale NoSQL databases. To serve their needs, we’re launching three new instance types in Limited Preview with up to 16 cores and 104 gigabytes of RAM. They are available in the familiar standard, high-memory and high-CPU shapes.