Elizabeth
My wife and I had to go to Elizabeth, NJ today. Contrary to its reputation and Google street view, it looked quite nice.285 N. Broad St., Elizabeth, NJSure, there were some sketchy parts off the main...
View ArticleNetra T1 105
I accidentally became the owner of a new old Sun Netra T1 model 105 yesterday. I had asked our super-helpful systems administrator if he had any old 1U servers to loan me as a prop for a tabletop demo...
View ArticleSpringtime
It is starting to look a little like springtime on campus. Some (but certainly not all!) trees are starting to bud......and the groundhogs that live in the courtyard between the Engineering A and D...
View ArticleJumping Spider
A little jumping spider was jumping along the shelf at my desk today, so I figured it might be a great time to test out the macro mode on our new camera. I learned a few things as a result:macro mode...
View ArticleDepressing view from work
My desk doesn't really have the best window view in the world. For almost five years now (since January 2008), this is the view I've seen out my window:The big contraptions are a pair of Strobic Air...
View ArticlePolitical Climates
Just a brief thought here--hopefully I don't land on some watch list for this.I hang out on FreeNode during my workday and one of my favorite channels is one about high-performance computing (or...
View ArticleOptiPlex 990 guts
I took a few photos of one of the Dell OptiPlex 990 business desktops I recently installed in our lab. They're pretty standard as far as desktops go, but some of the pictures came out nicely.Here's...
View ArticleDisk failure: Linux vs UNIX
I woke up this morning to a dozen emails from OSSEC, all saying the same thing:smartd[6835]: Device: /dev/sdg, SMART Failure: FAILURE PREDICTION THRESHOLD EXCEEDEDThis wasn't entirely unanticipated;...
View ArticleGetting a soda
Walking down the hall outside my labGoing towards the walkway connecting A-wing and B-wingView from the 2nd floor of the walkway connecting A- and B-wingsWalking down the staircase at the back of...
View ArticleBlue Gene/P First Impressions - Hardware
The institution at which I work recently acquired a Blue Gene/P (BG/P) system as the first of three phases of building an advanced supercomputing center here, and as a computationalist in materials...
View ArticlePlaying with OpenMP in my research
The core simulation code our research group has been using for the last two decades has remained relatively static, with the only major increases in speed coming from running it on faster processors....
View ArticleBraindead thread scheduling in Linux
In my previous post I showed some of the scaling benchmarks I was getting with my newly multithreaded molecular dynamics code. Those scaling figures were done on a dedicated node though, and I found...
View ArticleMassive Parallelism and the Outlook for CUDA
Massive ParallelismThe future of high performance computing (and the so-called exascale computing) lies squarely in the realm of massively parallel computing systems. When I first began hearing about...
View ArticleQuick MPI Cluster Setup on Amazon EC2
PremiseI was recently tasked with doing some MPI benchmarks on Amazon EC2 to get a general idea of how well EC2's Cluster Compute capabilities perform. I had never used EC2 before, so the notion of...
View ArticleMPI Benchmarks: Amazon EC2 Cluster Compute vs. Myrinet 10Gig
As a followup to my previous post on how to set up MPI across Amazon's Cluster Compute nodes, here are the results I got from OSU's microbenchmarks. I ran each point-to-point test five times, and the...
View ArticleDual-Rail QDR as an Alternative to FDR Infiniband
The data and opinions expressed here are solely my own and do not reflect those of my employer or the providers of the computing resources used.Fourteen Data Rate (FDR) is the latest iteration of...
View ArticleBuilding IPM 0.983 for lightweight MPI profiling
IPM, or Integrated Performance Monitoring, is a lightweight library that takes advantage of the MPI standard's builtin profiling interface which actually positions the standard user-facing MPI routines...
View ArticleReality Check on Cloud Usage for HPC
The opinions and analysis expressed here are solely my own and do not reflect those of my employer or the National Science Foundation.I get very bent out of shape when people start speaking...
View ArticleWhat's Killing Cloud Interconnect Performance
A small blog post benchmarking MPI on Google's new Google Compute Engine crossed my path this weekend, and since I've been messing around with virtualized interconnect performance lately, it caught my...
View ArticleHadoop for Real Science
About a year ago I developed a cursory understanding of data-intensive compute technologies like Hadoop while attending VSCSE's Science Cloud Summer School, but I never bothered putting much time into...
View Article