staff blogs

distributed.net staff keep (relatively) up-to-date logs of their activities in .plan files. These were traditionally available via finger, but we've put them on the web for easier consumption.

2019-08-23

August 2019 update

Dear friends,

Great news – the OGR-28 project has now been progressing for 2000 days and has brought 60,000 people from around the world together working towards a single goal. If you have recently joined us, welcome :)

The combined effort of our group using commodity computer equipment is producing the same amount of work that would be produced by a major supercomputer – such as the Cray XC40 model used in Texas Advanced Computer Center’s Lonestar 5. A single Cray XC40 cabinet contains 192 nodes, each with 24 cores of Intel Xeon CPU power and costs upwards of $500,000; with a power cost of about $100,000 per year.

We know that some of you have been with us since the beginning, and we are grateful for your contributions. We’ve come such a long way since 1997. A Pentium 100 computer could calculate 100,000 keys/sec of RC5. A $150 graphics card today is 30,000 times faster. Our first OGR project searching for the optimal Golomb ruler with 24 marks took a little over 4 years to complete. At the time we started OGR-24, OGR-28 looked almost impossible. Thanks to the sterling efforts of our brilliant coders, and you, our volunteers, we have proved optimal rulers for 25, 26 and 27 marks. We’re well on our way to proving the optimal ruler with 28 marks, where our throughput is 100 times faster than when we started on OGR-24. Even as we approach 70% completion on the project, there is still a better than 1 in 7 chance that the OGR stub your computer works on today has not been calculated before.

One way you can help is to download the latest version of the client software for your platform. We will be releasing new clients shortly which can support the latest processors and graphics cards from Intel, AMD and ARM. Whether you have a Core, Ryzen, Pi, Galaxy or Xeon (including Xeon Phi), we have a client for you.

Following on from her excellent work on RC5 ARM64 cores, our good friend Sunset has written a core for processors featuring AVX-512 instructions which is an astounding 70% quicker than the cores we have, which are already highly optimised.

Telling your friends about distributed.net and our projects – and working together with them in a team – will help us take months and years off our expected project completion dates. Perhaps you have a friend who used to run dnetc in our early days? Let them know that we are still here :)

Moo!

2018-01-08

New Year 2018

Filed under: clients,project status,stats — Mike Reed @ 23:12 +00:00

Dear friends,

We hope that you celebrated a happy holiday season and saw the new year in with friends and family wherever you may be.

Another year passes… but we press on. OGR-28 is almost in touching distance. Perhaps by the end of next year, we will have completed it. You can help bring that day closer by telling your friends to help us out. :)

We have made great progress during the year by working together. Friends and strangers, enthusiasts all, from the four corners of the globe in pursuit of a common goal.

On the OGR-28 project, we are making great strides through the second pass, with more than 387 million stubs (out of 524 million) checked once and 50 million verified. We recently re-opened the 8th stub space.

On the RC5-72 project, we check more than 21 million blocks every day. Perhaps we shall reach 5% of all work completed within a month. Graphics cards now exist which can process 5 Gigakeys per second. For comparison, the overall rate of our network back when we were working on the RC5-64 project was 102 Gigakeys per second. Our current network speed processing RC5-72 is over 1.2 Terakeys per second!

We continue to look for ways to optimise our client software; some of the newer GPU clients are increasingly I/O-bound, so we will be releasing new versions with increased buffer sizes for them shortly.

Our friend Sunset has recently created a new core for RC5-72 on ARM64 which is almost 5 times quicker than our existing core. If you have a Raspberry Pi 3 or a Cavium Thunder X2 system, you can make use of it.

We are always on the look out for new projects which we can lend our expertise and experience to. If you have a project in mind or some code to hand, please let us know.

We already experienced our first “megaflush” of the year; 6 million blocks in one day. We like to encourage participation, but would like to remind everyone that holding blocks for extended periods of time can cause wasted, duplicated work to be done by other participants if we happen to need to recycle work for another pass.

Moo!

2017-03-11

A quick update

Filed under: Uncategorized — Mike Reed @ 18:44 +00:00

We have opened the 22/x stub space on OGR-28 this weekend. In a little over 3 years, we have completed more than 29% of the work necessary to prove the Optimal Golomb Ruler with 28 marks. We have also just passed the ‘300 million stubs completed’ milestone. Over 47,700 of us working together have helped to make this happen.

One of our newest contributors on the coding side, Sunset, has helped us make a native PPC64 client for systems running Linux. She has some experience writing for VMX (which you may know as Altivec) and is hoping to update our existing cores for 64 bit systems. If you have a Power9 system in action, we would love to hear from you.

Moo!

2016-08-10

Quick update – 10 August 2016

Filed under: clients,project status,stats — Mike Reed @ 02:18 +00:00

Apologies, it has been a while since I published a blog update. The front page of our web site still has a problem, but we are making progress on our projects. A look at OGR-28: 42,000 participants. 900 days. 31 BILLION (I know!) Gnodes checked. 20% completed. We can still complete it by New Year if each of us gets a few friends to join in!

RC5-72 has been in progress for 5,000 days as I write. There is still a long way to go, with only just over 4% of the work completed. To have the help of 120,000 participants is amazing. Thank you for all of your contributions.

Have you downloaded new dnetc client software lately? There are a raft of improvements in the last couple of versions, including speed-ups on some platforms (the latest came out of pre-release two months ago). Moo!

2016-02-23

2 year OGR-28 anniversary update!

Filed under: project status,stats — Mike Reed @ 22:43 +00:00

Dear friends,

We’re happy to announce a new client software version, 2.9112.521, for all major operating systems. These are available on the pre-release page: https://www.distributed.net/Download_prerelease. We would be happy for you to check them out and give us your feedback! Among the improvements are better GPU support, OpenCL on more platforms, and a new AVX2 core for RC5-72 contributed by Yasuhiro Katsui which runs 3.6x faster than the previous code. If you find any issues, please report them to us at https://bugs.distributed.net/. We expect to move the build 521 clients to release status after a short time.

As of this writing, we have just passed the second anniversary of the OGR-28 project. With over 36,000 contributors from around the world, we have completed more than 14% of the work necessary to prove the Optimal Golomb Ruler with 28 marks. You can track our progress on https://stats.distributed.net/, where we’ve recently restored the popular keyrate graphs, which show changes in our network’s speed over time. This one is for OGR-28: https://stats.distributed.net/keyrate.php?project_id=28

As you might have noticed above, many of our websites are also now available via https. Feel free to update your bookmarks.

We are grateful for all of your contributions. Please tell your friends about our projects, and watch those keyrate graphs go up!

Moo!

2016-01-20

Quick update – 20 January 2016

Filed under: project status — Mike Reed @ 22:34 +00:00

Dear friends,

This week we celebrated the birthday of our project leader Bovine… and 700 days of searching for the optimal Golomb Ruler with 28 marks! With your help, we have completed 13.61% of the entire work space, and the first 7 hills. (See my earlier blog posts!)

We also learned today that our friends at the Great Internet Mersenne Prime Search (mersenne.org) have discovered the largest prime number known to mankind to date!

Keep telling your friends :) Moo!

2016-01-12

Quick update – January 12 2016

Filed under: project status,stats — Mike Reed @ 23:52 +00:00

Dear friends,

With nearly 35,000 volunteers and 13.38% of the search space completed, we’ve just started on a new OGR range today (8-x) and so daily stubs completed will be higher. We’re still struggling with our web site, but glad that you’re here with us. Keep telling your friends. We need all the help we can get. Moo!

2015-12-31

Happy New Year

Filed under: project status — Mike Reed @ 19:23 +00:00

Happy new year to you from all of us at distributed.net. We are grateful for all of your contributions over the past year and look forward to working with you in the new year as we try to prove the optimal Golomb ruler with 28 marks. Moo!

2015-12-04

Quick update – December 4 2015

Filed under: project status — Mike Reed @ 20:48 +00:00

Our web site has a plain appearance for a few days. Bovine is updating the software that drives its appearance. Please don’t be alarmed and carry on contributing as normal. :) Moo!

2015-12-02

Quick update – December 2 2015

Filed under: project status — Mike Reed @ 21:38 +00:00

Dear friends,

We’ve recently started on a new OGR range (7-x) and so daily stubs completed will be higher. We’re still struggling with our web site, but glad that you’re here with us. Keep telling your friends. We need all the help we can get. Moo!