Re: [gmx-users] Optimal Hardware for Gromacs

2010-05-02 Thread David van der Spoel
On 2010-05-02 10.32, Alexey Shvetsov wrote: Hi, Actualy suppermicro offers 4 socket machines with 12-core amd cpus so you get 48cores in 1U GROMACS perform well on such hardware =) Sure but a simple rule of thumb is that you want to maximize aggregate clockspeed per dollar. Such machines

Re: [gmx-users] Optimal Hardware for Gromacs

2010-05-02 Thread Alexey Shvetsov
2010/5/2 David van der Spoel sp...@xray.bmc.uu.se On 2010-05-02 10.32, Alexey Shvetsov wrote: Hi, Actualy suppermicro offers 4 socket machines with 12-core amd cpus so you get 48cores in 1U GROMACS perform well on such hardware =) Sure but a simple rule of thumb is that you want to

[gmx-users] Optimal Hardware for Gromacs

2010-04-28 Thread Igor Leontyev
Hi, there is a question for hardware experts. What is the optimal hardware to achieve better scalability in parallel MD simulations of 10K-100K atoms? Probably, it should be a cluster of multi-cpu, multi-core units with fast interconnection. In this case, what is the optimal=performance/price

Re: [gmx-users] Optimal Hardware for Gromacs

2010-04-28 Thread David van der Spoel
On 4/28/10 9:38 AM, Igor Leontyev wrote: Hi, there is a question for hardware experts. What is the optimal hardware to achieve better scalability in parallel MD simulations of 10K-100K atoms? Probably, it should be a cluster of multi-cpu, multi-core units with fast interconnection. In this