Profile

Cover photo
Murray Stokely
Works at Google
Attended Oxford University
Lives in San Francisco
585 followers|580,169 views
AboutPostsPhotosYouTubeReviews

Stream

Murray Stokely

Shared publicly  - 
 
 
In which John McAfee joins the list of people who answer the question, "What do people use to get stuff done?".
2
1
Sidney Lucas's profile photo
Add a comment...

Murray Stokely

Shared publicly  - 
 
Pre-print of our article on RProtoBuf #rstats  is up on arXiv.  Thanks +Dirk Eddelbuettel and +Jeroen Ooms for work on the writeup and +Romain Francois for much of the code.
 
Protocol Buffers are a popular method of serializing structured data between applications - while remaining independent of programming languages or operating systems. They offer a unique combination of features, performance, and maturity that seems particularly well suited for data-driven applications and numerical computing. 
4
Add a comment...

Murray Stokely

Shared publicly  - 
 
 
HotCloud '14 CFP now open: http://ow.ly/sy0g8 6th USENIX Workshop on Hot Topics in Cloud Computing
1
Add a comment...

Murray Stokely

Shared publicly  - 
 
Well done +Daniel Belov et al!  Cover of Nature Chemistry, not bad. =)
1
Add a comment...

Murray Stokely

Shared publicly  - 
 
Considering writing a short position paper about distributed storage systems.  5-page limit in HotStorage or 6-page short paper in +USENIX ATC.  Decisions, decisions..  Any thoughts?
1
Adnan Hassan's profile photo
 
Hot storage for sure !
Add a comment...
In his circles
334 people
Have him in circles
585 people
Denise Snow's profile photo
Abi Talk's profile photo

Murray Stokely

Shared publicly  - 
 
 
"""
The stagnation of the Ming may carry important lessons for a more modern superpower: The United States. We too are a huge, rich, powerful nation that for much of our history has dominated the field of competitors. We too have a whole century of dominance — the 20th — under our belt. And if there's one thing we don't want to do, it's turn into the Ming.
"""
1
Nicholas Miceli's profile photo
 
The Merciless?
Add a comment...

Murray Stokely

Shared publicly  - 
 
 
When you freeze water, it expands -- this is an important and unusual property of water, since most substances are denser as solids than as liquids, and so a given weight of them is smaller when it solidifies. In fact, this tendency to expand is remarkably forceful, and if you try to freeze water in a box which doesn't give it room to expand, you're going to need a new box.

No, really. This article tells you about some of the experiments people tried with this (e.g., sealing them in artillery shells; when the water finally froze, their cast-iron plugs ended up flying away at remarkable speed). The reason it doesn't, ultimately, work, is that ordinary ice -- what's technically known as "Ice-1" -- is just plain bigger than water, and the only way to make it smaller is to put enough pressure on it that the molecular structure of the ice changes. At a pressure of 1,736 atmospheres, it transforms into "Ice-II," which is denser than water: instead of forming hexagonal cells (the same hexagons which lead ordinary ice to form hexagonal snowflakes), the water molecules now arrange themselves into slanted rectangular prisms. 

It turns out that ice actually has more than a dozen such forms, each with different physical and chemical properties; Ice-XV, for example, resists holding any kind of electrical charge, and the as-yet-uncreated (and unnamed) forms of ice above a pressure of 10 million atmospheres would be metallic. 

h/t +Alex Scrivener and +Keith Wilson for the link.
2
Add a comment...

Murray Stokely

Shared publicly  - 
 
CFP: Workshop on Job Scheduling Strategies for Parallel Processing (JSSPP) at @IEEE #IPDPS  2014.  Deadline in 2 weeks: http://www.cs.huji.ac.il/~feit/parsched/jsspp14/

The JSSPP workshop addresses all scheduling aspects of parallel processing.

Large parallel systems have been in production for about 20 years, creating the need of scheduling for such systems. This workshop was created in 1995 to provide a forum for the research and engineering community working in the area. Initially, parallel systems were very static. Machines were built in fixed configurations, which would be wholesale replaced every few years. Much of the workload consisted of parallel scientific jobs. These jobs were static, running on a fixed number of nodes. Systems were primarily managed via batch queues. The user experience was far from interactive; jobs could wait in queues for days or even weeks.

A little over 10 years ago, the emergence of large scale, interactive, web applications began to drive the development of a new class of systems and schedulers. These systems would run “services”, which would essentially never terminate (unlike scientific jobs). This created systems and schedulers with vastly different properties. Moreover, this created an enormous demand for computing resources, resulting in a commercial market of competing providers. At the same time, the increasing demands for more power and interactivity have driven scientific platforms in a similar direction, causing the lines between these platforms to blur.

Nowadays, parallel processing is much more dynamic and connected. Many workloads are interactive and make use of variable resources over time. Complex parallel infrastructures can now be built on the fly, using resources from different sources, provided with different prices and quality of services. Capacity planning became more proactive, where resources are acquired continuously, with the goal of staying ahead of demand. The interaction model between job and resource manager is shifting to one of negotiation, where they agree on resources, price, and quality of service. These are just a few examples of the open issues facing our field.

JSSPP solicits papers that address any of the challenges in parallel scheduling, including:

Design and evaluation of new scheduling approaches.
Performance evaluation of scheduling approaches, including methodology, benchmarks, and metrics.
Workloads, including characterization, classification, and modeling.
Consideration of additional constraints in scheduling systems, like job priorities, price, accounting, load estimation, and quality of service guarantees.
Impact of scheduling strategies on application performance, user friendliness, cost efficiency, and energy efficiency.
Scaling and composition of very large scheduling systems.
Cloud provider issues: capacity planning, service level assurance, reliability.
Interaction between schedulers on different levels, like processor level as well as whole single- or even multi-owner systems.
Experience reports from production systems.
Experience reports from large scale compute campaigns.
From its very beginning, JSSPP has strived to balance practice and theory in its program. This combination provides a rich environment for technical debate about scheduling approaches including both academic researchers as well as participants from industry. JSSPP is a high-visibility workshop, which has been ranking repeatedly in the top 10% of Citeseer’s venue impact list.
1
Add a comment...

Murray Stokely

Shared publicly  - 
 
I released a new version of the HistogramTools #rstats package including new Histogram distance measures and other functions that have come up in my work in analyzing large datasets.
A new version 0.3 of HistogramTools is now on CRAN. HistogramTools provides a number of methods for manipulating histograms, measuring the distance between histograms, calculating the information loss due to binning aggregate...
1
Ahmed Ali's profile photo
 
Great tool!
Add a comment...

Murray Stokely

Shared publicly  - 
 
Pushed a new version of HistogramTools to CRAN today.  This version adds some new distance measures for histograms, a few new visualizations, and improved vignettes.  Thanks +Tim Hesterberg for the thorough code review.  More updates and some writeups about applications are coming next year.  #rstats http://cran.r-project.org/web/packages/HistogramTools/
1
Add a comment...
People
In his circles
334 people
Have him in circles
585 people
Denise Snow's profile photo
Abi Talk's profile photo
Education
  • Oxford University
  • California State University
  • Independent University of Moscow
Basic Information
Gender
Male
Work
Occupation
Software Engineer
Employment
  • Google
    Software Engineer, present
  • Windriver
  • BSDi
  • Walnut Creek CDROM
Places
Map of the places this user has livedMap of the places this user has livedMap of the places this user has lived
Currently
San Francisco
Previously
Florida, USA - San Francisco Bay Area, California - Moscow, Russia - Shanghai, China - Oxford, UK
This hotel is in a nice location, but we had a horrible experience here straight from the start when we arrived at 6pm to find that our 2 bedroom family suite that we booked/confirmed/and had printed documentation for was not available and instead they had decided to move us to a large single bedroom. The staff was completely unapologetic and let it slip that they could get 400 euros a night for each of the two bedrooms of our 2-bedroom family suite, which was more than what we had booked for. After making us wait around for a long time while calling the elusive manager, they tried to show us a slightly larger single bedroom (meant for 3, not 5) which again was not half as big as the 2-bedrooms we needed, and only finally relented in giving us two rooms non-adjoining that we could use for the first night. However, they then made us check out the next day and check back into a smoking 2-bedroom family suite when we had reserved non-smoking. The rooms were disgusting as one smelled strongly of human excrement and the other had an incredibly strong stench of smoke. The front desk also just goes out of its way to not be helpful and they charge for even really small things like 1-2 euros charged to your room for calling a taxi on your behalf, etc. When we checked out they had the temerity to ask for payment of the full amount of the rooms we booked despite booking us in much smaller, smoking rooms despite our reservation for a larger non-smoking room.
• • •
Public - a year ago
reviewed a year ago
Great park and indoor recreational facilities. Gym, basketball courts, lots of activities for kids, summer camps, easter egg hunts, etc..
Public - 2 years ago
reviewed 2 years ago
Same day delivery, reasonable prices, knowledgeable staff. Limited selection.
Public - 3 years ago
reviewed 3 years ago
4 reviews
Map
Map
Map
Public - 3 years ago
reviewed 3 years ago