The conventional gravity-geologic method (GGM), as a widely used method for bathymetric prediction, is a single-point calculation method between the gravity anomaly and unknown bathymetry. To enhance the accuracy of bathymetric prediction, the improved GGM (IGGM) reliably removing the long-wavelength gravity effect of regional seafloor topography was proposed. The modeling of the long-wavelength gravity field was refined by calculating the short-wavelength gravity correction at the control points based on the weight parameters introduced into the Bouguer slab formulation. The IGGM bathymetric model for the experimental area (113°E-119°E, 12°N-19°N) in the South China Sea was constructed by combining shipborne bathymetry data from the National Centers for Environmental Information (NCEI) and the V31.1 gravity anomaly model from the Scripps Institution of Oceanography (SIO). The standard deviation of the difference between the IGGM model and shipborne bathymetry was approximately 103.46 m at iteration points, which was better than the DTU18, GEBCO_2021 and topo_23.1 bathymetric models. Compared with the accuracy of the model derived from the GGM, the accuracy of the model from IGGM was improved by about 27%. At checkpoints, the accuracy of IGGM was improved by approximately 17 m, and the improvement was larger in areas with complex terrain.

Virtualization has become an important technology that is used across many platforms, particularly servers, to increase utilization, multi-tenancy and security. Virtualization introduces additional overhead that often relates to memory management, interrupt handling and hypervisor mode switching. Among those, memory management and translation lookaside buffer (TLB) management have been shown to have a significant impact on the performance of systems. Two principal mechanisms for TLB management exist in today's systems, namely software and hardware managed TLBs. In this paper, we analyze and quantify the overhead of a pure software virtualization that is implemented over a software managed TLB. We then describe our design of hardware extensions to support virtualization in systems with software managed TLBs to remove the most dominant overheads. These extensions were implemented in the Power embedded A2 core, which is used in the PowerEN and in the Blue Gene/Q processors. They were used to implement a KVM port. We evaluate each of these hardware extensions to determine their overall contributions to performance and efficiency. Collectively these extensions demonstrate an average improvement of 232% over a pure software implementation.

In many computing domains, hardware accelerators can improve throughput and lower power consumption, instead of executing functionally equivalent software on the general-purpose micro-processors cores. While hardware accelerators often are stateless, network processing exemplifies the need for stateful hardware acceleration. The packet oriented streaming nature of current networks enables data processing as soon as packets arrive rather than when the data of the whole network flow is available. Due to the concurrence of many flows, an accelerator must maintain and switch contexts between many states of the various accelerated streams embodied in the flows, which increases overhead associated with acceleration. We propose and evaluate dynamic reordering of requests of different accelerated streams in a hybrid on-chip/memory based request queue in order to reduce the associated overhead.

The transistor density of microprocessors continues to increase as technology scales. Microprocessors designers have taken advantage of the increased transistors by integrating a significant number of cores onto a single die. However, a large number of cores are met with diminishing returns due to software and hardware scalability issues and hence designers have started integrating on-chip special-purpose logic units (i.e., accelerators) that were previously available as PCI-attached units. It is anticipated that more accelerators will be integrated on-chip due to the increasing abundance of transistors and the fact that not all logic can be powered at all times due to power budget limits. Thus, on-chip accelerator architectures deserve more attention from the research community. There is a wide spectrum of research opportunities for design and optimization of accelerators. This paper attempts to bring out some insights by studying the data access streams of on-chip accelerators that hopefully foster some future research in this area. Specifically, this paper uses a few simple case studies to show some of the common characteristics of the data streams introduced by on-chip accelerators, discusses challenges and opportunities in exploiting these characteristics to optimize the power and performance of accelerators, and then analyzes the effectiveness of some simple optimizing extensions proposed.

Clock gating is a well-known technique to reduce chip dynamic power. This paper analyzes the disadvantages of some recent clock gating techniques and points out that they are difficult in system-on-chip (SoC) design. Based on the analysis of the intellectual property (IP) core model, an adaptive clock gating (ACG) technique which can be easily realized is introduced for the low power IP core design. ACG can automatically enable or disable the IP clock to reduce not only dynamic power but also leakage power with power gating technique. The experimental results on some IP cores in a real SoC show an average of 62.2% dynamic power reduction and 70.9% leakage power reduction without virtually performance impact.

This paper presents a new technique eliminating redundant cache tag and data accesses to reduce energy consumption. We assign a register to each tag in a cache to represent its state. Before starting an access, we can check the tag states in the target cache set to determine which way(s) should be accessed and which should not. Through this method, almost all the accesses in the I-cache can be directed to the target cache way immediately for most benchmark programs. For a 2-way set-associative cache, the energy consumption can be reduced by 76.6% compared with conventional cache architecture, and by 39.8% compared with Block Buffering, a simple but well-known technique. Besides, this approach does not require any special circuitry internal to the cache RAM such as row or column activation mechanisms. This is considered an important advantage in industry because of its easy implementation.