Sign In
Register
Help
Post your comments for Sun's Rock Doomed From Start, Analysts Say here
Page 1 of 1
Sun's Rock Doomed From Start, Analysts Say
MultiQuote
#2
DavidHalko 
- Newbie
-
- Group: Members
- Posts: 1
- Joined: 04-July 09
Posted 04 July 2009 - 01:11 PM
"Earlier this week, the New York Times reported that Sun had cut development of the 16-core Rock chips"
The New York Times reported unconfirmed statements from two unnamed people. The rumor may be correct (considering the silence on the topic from Sun) - but the New York Times did not report a confirmed statement of Sun cutting development.
"the 16-core Rock chips, which were designed to go into high-end servers."
The 'Rock' UltraSPARC RK processors were designed around the mid-range 'SuperNova' Servers, speculated to be code-named: AT7180, AT7280, AT7480, AT7880 from OpenSolaris source-code tree, possibly targeting 1-8 sockets.
http://www.theregist.../2009/02/04/sunsupernovalineup/
The 'Niagra' UltraSPARC T1 and T2 are used in low-end servers with 1 socket.
The 'Niagra' UltraSPARC T2+ tops out the low-end and brushes the mid-range servers with 2-4 sockets.
The SPARC64 VI, VII are used in mid-range to high-end 'Advanced Product Line' servers using 1-64 sockets.
'Rock' based 'Super Nova' server 1-8 socket count appears to be smaller than SPARC64 based 'Advanced Product Line' server 1-64 socket count, yet double UltraSPARC T+ 'Niagra' based server 2-4 socket count.
This places 'Rock' squarely in the mid-range - not in the high-end.
"...Sun's fastest server processor today, the eight-core UltraSparc T2"... "was targeted at enterprise servers to process data-intensive applications like databases."
The 'Advanced Processor Line' based upon SPARC64 (and 'Super Nova' line based upon 'Rock' or UltraSPARC RK) are designed for data-facing applications such as database applications.
The 'Niagra' UltraSPARC T2 processor (with embedded 10GigE and octal crypto engines) is designed for network-facing applications, especially where encryption is needed. The T2 has been the fastest and unchallenged single socket web server on the market from 2007 through and including today (mid-2009.)
http://www.spec.org/...3=-1&format=tab
Sun indicates the ideal applications for the T2 based servers include: single socket Blade & Rack Servers, Wireless Infrastructure, Telco Infrastructure.
http://www.sun.com/p...ecs.xml#anchor3
One would expect basic technology mistakes from the New York Times, whose publication is not aimed at a computer savvy readership.
It is very reasonable for computer savvy publications to pass on to their readership stories broken by other publications.
To expand upon statements by a referenced publisher, to alluded they said something they did not, is bad practice.
One would also normally expect publications targeting computer savvy readership to fix incorrect statements cited by other publications, rather than just propagate inaccurate information.
The New York Times reported unconfirmed statements from two unnamed people. The rumor may be correct (considering the silence on the topic from Sun) - but the New York Times did not report a confirmed statement of Sun cutting development.
"the 16-core Rock chips, which were designed to go into high-end servers."
The 'Rock' UltraSPARC RK processors were designed around the mid-range 'SuperNova' Servers, speculated to be code-named: AT7180, AT7280, AT7480, AT7880 from OpenSolaris source-code tree, possibly targeting 1-8 sockets.
http://www.theregist.../2009/02/04/sunsupernovalineup/
The 'Niagra' UltraSPARC T1 and T2 are used in low-end servers with 1 socket.
The 'Niagra' UltraSPARC T2+ tops out the low-end and brushes the mid-range servers with 2-4 sockets.
The SPARC64 VI, VII are used in mid-range to high-end 'Advanced Product Line' servers using 1-64 sockets.
'Rock' based 'Super Nova' server 1-8 socket count appears to be smaller than SPARC64 based 'Advanced Product Line' server 1-64 socket count, yet double UltraSPARC T+ 'Niagra' based server 2-4 socket count.
This places 'Rock' squarely in the mid-range - not in the high-end.
"...Sun's fastest server processor today, the eight-core UltraSparc T2"... "was targeted at enterprise servers to process data-intensive applications like databases."
The 'Advanced Processor Line' based upon SPARC64 (and 'Super Nova' line based upon 'Rock' or UltraSPARC RK) are designed for data-facing applications such as database applications.
The 'Niagra' UltraSPARC T2 processor (with embedded 10GigE and octal crypto engines) is designed for network-facing applications, especially where encryption is needed. The T2 has been the fastest and unchallenged single socket web server on the market from 2007 through and including today (mid-2009.)
http://www.spec.org/...3=-1&format=tab
Sun indicates the ideal applications for the T2 based servers include: single socket Blade & Rack Servers, Wireless Infrastructure, Telco Infrastructure.
http://www.sun.com/p...ecs.xml#anchor3
One would expect basic technology mistakes from the New York Times, whose publication is not aimed at a computer savvy readership.
It is very reasonable for computer savvy publications to pass on to their readership stories broken by other publications.
To expand upon statements by a referenced publisher, to alluded they said something they did not, is bad practice.
One would also normally expect publications targeting computer savvy readership to fix incorrect statements cited by other publications, rather than just propagate inaccurate information.
Page 1 of 1
