PARS: A scheduling of periodically active rank to optimize power efficiency for main memory

• Motivated by the increasing power consumption of the main memory, while its growing significantly in both speed and capacity for it is a major shared resource among cores in a multi-core system. In this paper, we propose a periodically active rank scheduling (PARS) to optimize power efficiency for DRAM in smart phones.
• We coordinate page allocation policy with operating system scheduler, which our page allocation policy make threads’ occupied memory rank more aggregation and regularly. PARS feature a three-step design.
• First, according to the relationship among threads, we partition all threads in the system into groups. For example, threads belonging to the same process are sharing a group. Moreover, combining both priority balance and load balance, the number of thread groups is equal to the number of the memory ranks.
• Thread group and memory rank are one-to-one correspondence. Second, according to the group number the thread belonging, allocate page in corresponding memory rank but in different banks. All occupied memory of threads in a group aggregates into one corresponding memory rank.
• Finally, priority schedule threads in the same group, after all threads in the same group have run sometime, choosing another. When a group is running, only its memory rank is active, after choosing another, switching active memory rank simultaneously. So, memory rank periodically active which can optimize power efficiency without reducing much performance from migrating.

The growing gap between microprocessor speed and DRAM speed is a major problem that computer designers are facing. In order to narrow the gap, main memory is expected to grow significantly in both speed and capacity for it is a major shared resource among cores in a multi-core system, which will lead to increasing power consumption, perhaps replacing the dominant consuming fraction of processor. Therefore, it is critical to address the power issue without seriously decreasing performance in the main memory subsystem. In this paper, we propose a periodically active rank scheduling (PARS) to optimize power efficiency for DRAM in smart phones. Our scheduling features a three-step design. First, we partition all threads in the system into groups. Second, we modify page allocation policy to achieve threads in the same group which occupies the same rank but different banks of DRAM. Finally, we sequentially schedule threads in one group after another while only active running group׳s ranks to retain other ranks low power status. As a result, our scheduling periodically activates one rank after another to optimize memory power efficiency. We implement PARS in Linux 2.6.39 kernel running randomly generated workloads containing single-threaded and multi-threaded benchmark. Experimental results show that PARS can improve both the memory power efficiency by 26.8% and performance by 4.2% average relative to default system while reducing negligible fairness.