Submission + - Nexus 4 includes support for LTE (anandtech.com)
It seems that a simple software update can allow the Nexus 4 smartphone to run on LTE Band 4. All users have to do is dial *#*#4636#*#* (INFO) or launch the Phone Info app. After that, choosing to connect to AWS networks should allow the Nexus 4 to run on LTE networks on Band 4.
The AnandTech report states explicitly that the LG Nexus 4 only works on LTE Band 4, on 1700/2100MHz frequencies, and supports bandwidths of 5,10, and 20MHz.
Submission + - Scientists See Advances in Deep Learning, a Part of Artificial Intelligence - NY (nytimes.com)
Submission + - What Nobody Tells You About Being a Game Dev (altdevblogaday.com)
Comment Re:Old problem (Score 1) 112
You can't checkpoint jobs at this scale. It will take longer to checkpoint a job then to compute an answer. This is further compounded when the job takes several months to run. A 1000 node cluster is very tiny compared to the scale they're talking about.
Comment Re:ummm, no. (Score 3) 112
Google is having the same problems that this article describes -- they haven't fixed it either.
If your problem domain can always be broken down into map-reduce, you can easily solve it with a hadoop-like environment to get fault tolerance. If your application falls outside of map-reduce (the applications this article is referring to), you need to start duplicating state (very expensive on systems of this scale) to recover from failures.
Comment Re:Old problem (Score 1) 112
How do you give the work to another node when the failed node contains the only copy of its state (like in an MPI job)? Duplicating the state on multiple nodes is way too expensive.
Comment Re:"and they halt operations when they do so" (Score 1) 112
Checkpoints will probably stick around for quite some time, but the model will need to change. Rather than serializing everything all the way down to a parallel filesystem, the data could potentially be checkpointed to a burst buffer (assuming a per-node design) or a nearby node (experimental SCR design). Of course, it's correct that even this won't scale to larger systems.
I think we'll probably have problems with getting data out to the nodes of the cluster before we start running into problems with checkpointing. The typical NFS home directory isn't going to scale. We'll need to switch over to something like udsl projections or another IO forwarding layer in the near future.
Comment Re:Into the Realm? (Score 1) 76
ANL's Mira is going to be roughly half as fast as LLNL's Sequoia.
Comment Re:10 years ago (Score 1) 231
If the developer is schizophrenic, then it might be n^2, otherwise it's probably (n(n-1))/2.
Comment So, this economist and... (Score 2, Insightful) 421
So, this economist and a computer scientist are sitting at a bar.... and these 5 girls walk in....
Behind the Scenes In Apple Vs. the Record Labels 146
Discovery Launch Delayed Due To Engine Issue 62
Man Robs Convenience Stores With Klingon "Batleth" 579