Machiko Asaie

Evaluation of the lock mechanism in a snooping cache

This paper discusses the design concepts of a lock mechanism for a Parallel Inference Machine (the PIM/c prototype) and investigates the performance of the mechanism in detail. Lock operations are extremely frequent on the PIM; however, lock contention rarely occurs during normal memory usage. For this reason, the lock mechanism is designed so as to minimize the lock overhead time in the case of no contention. This is done by using an invalidation lock mechanism, which utilizes the exclusive state of the snooping cache and in which the locked address is not broadcast. Experimental results demonstrate the benefits of the lock mechanism in regions of few lock contentions. They also confirm that, in most cases, the lock mechanism works well on the PIM. However, the mechanism is also found to cause performance degradation when a locked address is accessed by multiple processing elements (PEs) in a tightly-coupled multi-processor (TCMP). This is because shared data such as the flags for inter-PE communication, which are shared by all the PEs, may be accessed by multiple PEs at the same time, thus generating heavy contention. This paper also shows that combining a register-based broadcasting facility with the proposed lock mechanism can solve the above problem.

Evaluation of the Cluster Structure on the PIM/C Parallel Inference Machine

The characteristics of a cluster-structure parallel computer are analyzed and evaluated on the PIM/c parallel inference machine, which consists of eight-processor shared-memory clusters communicating through a processor connected to a network. To avoid communication bottlenecks, the maximum number of processors in a cluster is limited by the ratio of communication operations to program-execution operations. Since this ratio can be as high as 30% on the PIM/c, the network receiving operations should be distributed to processors in the same cluster.

Study on supporting technology for operational procedure design of IT systems in cloud-era datacenters

Datacenters are now widely used, and their sizes are increasing due to the rapid spread of cloud computing. Meanwhile, the cost of IT-system operations occupies 60% of the total cost in a datacenter. So far, improving operations has been focused on automating operations with operation management middleware. However, designing operational procedures, including composing and modifying operational manuals and checklists, is also a big issue since it is time consuming. Supporting technology that improves the efficiency of IT-system operational procedure design has been studied by focusing on the commonly used procedural parts. Existing checklists were analyzed to extract the iterated patterns of the procedures in order to evaluate the technology. The evaluation results showed that common operations such as the ones on management middleware and the remote log-in were extracted as procedural parts. The number of checklist composition steps was reduced by 45% on average and that of the modification steps was reduced by 87% on average by using these procedural parts. A prototyping tool was developed, and the checklists of a sample system different from the analyzed one were implemented on the tool.