GIMPS HITS {insert number here} COMPUTERS
 

Great Internet Mersenne Prime Search

PrimeNet is a world-wide, distributed Internet research computing system created in 1997 for the Great Internet Mersenne Prime Search (GIMPS). For further information, please email [email]primenet@mersenne.org[/email].

The virtual machine's sustained throughput* is currently 17456 billion floating point operations per second (gigaflops), or 1450.1 CPU years (Pentium 90Mhz) computing time per day. For the testing of Mersenne numbers, this is equivalent to 623 Cray T916 supercomputers, or 311.5 of Cray's most powerful T932 supercomputers, at peak power. As such, PrimeNet ranks among the most powerful computers in the world. (*Measured in calibrated P5 90Mhz, 32.98 MFLOP units: 25658999 FPO / 0.778s using 256k FFT.)

For more information, please see the GIMPS home page, the PrimeNet Statistics or the PrimeNet Project Credits.

Current PrimeNet Atomic Clock UTC Time is Tuesday 15 March 2005, 03:57:21
atom.gif (235 bytes)

Current Internet PrimeNet Server World Test Status
All Times are in Coordinated Universal Time (UTC)

These reports indicate the virtual supercomputer's operating status, automatically updated by the server every hour, on the hour. A complete table of the GIMPS overall project status is also available.

They consist of three sections: (1) system performance information and active exponent test ranges, (2) current assignments issued to GIMPS test clients, (3) exponents that have been eliminated as Mersenne prime candidates since the last master database synchronization. Section (1) does not include exponents in progress received from clients outside active IPS test ranges.

* Hourly World Test Status Summary
All Times are in Coordinated Universal Time (UTC)

This summary includes only section (1), and counts for sections (2) and (3).

Mersenne PrimeNet Server 4.0 (Build 4.0.031)
Status Summary Report 15 Mar 2005 03:00 (14 Mar 2005 19:00 Pacific)

This report is updated every 60 minutes
All dates and times are Coordinated Universal Time (UTC)
Assignment overdue check-in is set at 60.0 days (0.0 days to expire)


------- Aggregate CPU Statistics, P90 Units* -------

Last 7 Days Average Cumulative Today
from 08 Mar 2005 06h from 14 Mar 2005 06h
---------------------- ----------------------------------
Test Type CPU yr/day GFLOP/s CPU years CPU yr/day GFLOP/s
------------ ---------- ---------- ---------- ---------- ----------
Lucas-Lehmer 1405.406 16917.857 1339.370 1532.087 18442.799
Factoring 44.735 538.502 44.124 50.472 607.571
---------- ---------- ---------- ---------- ----------
TOTALS 1450.141 17456.359 1383.494 1582.559 19050.369



------- Internet CPU and Server Resources -------

Machines Applied on 55186 Accounts Server Synchronization 15 Feb 2005 03:47

Intel Pentium 4 : 27689 Total exponents merged : 393739
AMD Athlon : 36460 Updated only : 372253
Intel Pentium III : 5683 Added for testing : 0
Intel Pentium II : 1142 Retained in cleared list : 6765
Intel Celeron : 6701 Cleared tests removed : 14721
Intel Pentium Pro : 65 Manual tests removed / purged : 0
Intel Pentium : 403
AMD K6 : 388 Total Cleared by PrimeNet : 782504
Intel 486 : 20
Cyrix : 1311
Unspecified type : 168
---------------------- -------
TOTAL : 80030

how intresting even on [url]http://www.mersenneforum.org/showthread.php?t=3862[/url] ;) hehe we gots a lots goings ons

But when are we going to finally break that 18 Teraflop "barrier"?

I predict that we will break and sustain 18 tflops within the next 30 days. We have picked up 7315 computers in the last 30 days. They are split fairly 50/50 pentium/athlon. If alot are working on 10million digit numbers, results should start coming in over the next 30 day period. Since its about a 10% increase in computers we should see about a 10% increase in tflops. Hence that 18 tflop barrier shall fall soon... :banana:

Actually if we don't loose computers we should be able to sustain 18.5 tflops in the next 30-45 days

=====================================================

2/15 06h

Intel Pentium 4...............24851
AMD Athlon....................33134
Intel Pentium III...............5239
Intel Pentium II................1079
Intel Celeron...................6080
Intel Pentium Pro................61
Intel Pentium....................393
AMD K6...........................380
Intel 486...........................24
Cyrix.............................1311
Unspecified type...............163
---------------------- -------
TOTAL........................72715

======================================================


3/14 3:47

Intel Pentium 4...............27689
AMD Athlon...................36460
Intel Pentium III..............5683
Intel Pentium II...............1142
Intel Celeron...................6701
Intel Pentium Pro................65
Intel Pentium...................403
AMD K6..........................388
Intel 486..........................20
Cyrix............................1311
Unspecified type..............168
---------------------- -------
TOTAL........................80030

At Long Last!!!
 

:smile: :smile: :smile:

[url]www.mersenne.org/primenet[/url]

Great Internet Mersenne Prime Search

PrimeNet is a world-wide, distributed Internet research computing system created in 1997 for the Great Internet Mersenne Prime Search (GIMPS). For further information, please email [email]primenet@mersenne.org[/email].

The virtual machine's sustained throughput* is currently 18046 billion floating point operations per second (gigaflops), or 1499.2 CPU years (Pentium 90Mhz) computing time per day. For the testing of Mersenne numbers, this is equivalent to 644 Cray T916 supercomputers, or 322 of Cray's most powerful T932 supercomputers, at peak power. As such, PrimeNet ranks among the most powerful computers in the world. (*Measured in calibrated P5 90Mhz, 32.98 MFLOP units: 25658999 FPO / 0.778s using 256k FFT.)

For more information, please see the GIMPS home page, the PrimeNet Statistics or the PrimeNet Project Credits.

Current PrimeNet Atomic Clock UTC Time is Friday 18 March 2005, 19:37:40

[QUOTE]The virtual machine's sustained throughput* is currently 18046 billion floating point operations per second (gigaflops), [/QUOTE]

How many floating point operations is LL test of M(p) as a function of p?

Are trial factoring and P-1 factoring counted in total floating point operations?

correction



Machines Applied on 56349 Accounts Server Synchronization 15 Mar 2005 03:47

Intel Pentium 4 : 28333 Total exponents merged : 417296
AMD Athlon : 36871 Updated only : 398071
Intel Pentium III : 5796 Added for testing : 0
Intel Pentium II : 1132 Retained in cleared list : 6654
Intel Celeron : 7122 Cleared tests removed : 12571
Intel Pentium Pro : 71 Manual tests removed / purged : 0
Intel Pentium : 394
AMD K6 : 399 Total Cleared by PrimeNet : 795075
Intel 486 : 24
Cyrix : 1253
Unspecified type : 178
---------------------- -------
TOTAL : 81573


hehe
The virtual machine's sustained throughput* is currently 18990 billion floating point operations per second (gigaflops), or 1577.6 CPU years (Pentium 90Mhz) computing time per day. For the testing of Mersenne numbers, this is equivalent to 678 Cray T916 supercomputers, or 339 of Cray's most powerful T932 supercomputers, at peak power. As such, PrimeNet ranks among the most powerful computers in the world. (*Measured in calibrated P5 90Mhz, 32.98 MFLOP units: 25658999 FPO / 0.778s using 256k FFT.)

What a difference a few weeks make
Current PrimeNet Atomic Clock UTC Time is Friday 03 June 2005, 08:36:29
we've lost about 7000 computers in the last 8-10 weeks..... I guess its time to find another prime number... :cool:

Intel Pentium 4...........26518
AMD Athlon................32790
Intel Pentium III...........5233
Intel Pentium II............1065
Intel Celeron................7097
Intel Pentium Pro.............61
Intel Pentium................339
AMD K6........................345
Intel 486........................23
Cyrix............................949
Unspecified type............162
---------------------- -------
TOTAL.......................74582

IT SEAMS LIKE THE BIGGEST LOOSER IS AMD ATHLON DOWN BY ABOUT 4000

I guess a lot of people installed prime95 after the last prime discovery without really understanding what they were doing.

Discouraged by the long time it takes to do a single test, many uninstalled or forgot about the program.

[QUOTE=smh]
Discouraged by the long time it takes to do a single test, many uninstalled or forgot about the program.[/QUOTE]
I think you are right. That reminds us again that the server should assign DCs for every new user/machine, at least on his/her first assignment. This would serve the dual purpose of allowing the user to assess the reliability of the machine, and substantially reducing the time to obtain a first result. It would also help pushing forward the DC ranges, that are far behind the 1st timers.

I think one thing that could contribute to the reduction in computers is the end of the college year in April/May.
I think there are a number of computers that go offline during the summer (at least) that way.