Resource usage monitoring for KVM based virtual machines

Realization of cloud computing has been possible due to availability of virtualization technologies on commodity platforms. Measuring resource usage on the virtualized servers is difficult because of the fact that the performance counters used for resource accounting are not virtualized. Hence, many of the prevalent virtualization technologies like Xen, VMware, KVM etc., use host specific CPU usage monitoring, which is coarse grained. In this paper, we present a performance monitoring tool for KVM based virtualized machines, which measures the CPU overhead incurred by the hypervisor on behalf of the virtual machine along-with the CPU usage of virtual machine itself. This fine-grained resource usage information, provided by the above tool, can be used for diverse situations like resource provisioning to support performance associated QoS requirements, identification of bottlenecks during VM placements, resource profiling of applications in cloud environments, etc. We demonstrate a use case of this tool by measuring the performance of web-servers hosted on a KVM based virtualized server.

Automatic Data Allocation and Buffer Management for Multi-GPU Machines

Multi-GPU machines are being increasingly used in high-performance computing. Each GPU in such a machine has its own memory and does not share the address space either with the host CPU or other GPUs. Hence, applications utilizing multiple GPUs have to manually allocate and manage data on each GPU. Existing works that propose to automate data allocations for GPUs have limitations and inefficiencies in terms of allocation sizes, exploiting reuse, transfer costs, and scalability. We propose a scalable and fully automatic data allocation and buffer management scheme for affine loop nests on multi-GPU machines. We call it the Bounding-Box-based Memory Manager (BBMM). BBMM can perform at runtime, during standard set operations like union, intersection, and difference, finding subset and superset relations on hyperrectangular regions of array data (bounding boxes). It uses these operations along with some compiler assistance to identify, allocate, and manage data required by applications in terms of disjoint bounding boxes. This allows it to (1) allocate exactly or nearly as much data as is required by computations running on each GPU, (2) efficiently track buffer allocations and hence maximize data reuse across tiles and minimize data transfer overhead, and (3) and as a result, maximize utilization of the combined memory on multi-GPU machines. BBMM can work with any choice of parallelizing transformations, computation placement, and scheduling schemes, whether static or dynamic. Experiments run on a four-GPU machine with various scientific programs showed that BBMM reduces data allocations on each GPU by up to 75% compared to current allocation schemes, yields performance of at least 88% of manually written code, and allows excellent weak scaling.

A Brushless Wound Rotor Induction Generator for Variable Speed Microhydel Plants Without Ballast Load

A new generator topology for microhydel power plants, capable of unsupervised operation, is proposed. While conventional microhydel plants operate at constant speed with switched ballast loads, the proposed generator, based on the wound rotor induction machine, operates at variable speed and does away with the need for ballast loads. This increases reliability and substantially decreases system costs and setup times. The proposed generator has a simplified decoupled control structure with stator-referenced voltage control similar to a conventional synchronous generator, and rotor-side frequency control that is facilitated by rotating electronics mounted on the rotor. While this paper describes an isolated plant, the topology can also be tailored for distributed generation enabling conversion of the available hydraulic power into useful electrical power when the grid is present, and supplying local loads in the event of grid outage.

Two-dimensional Cartesian air-gap element (CAGE) for dynamic finite-element modeling of electrical machines with a flat air gap

This paper extends the air-gap element (AGE) to enable the modeling of flat air gaps. AGE is a macroelement originally proposed by Abdel-Razek et al.for modeling annular air gaps in electrical machines. The paper presents the theory of the new macroelement and explains its implementation within a time-stepped finite-element (FE) code. It validates the solution produced by the new macroelement by comparing it with that obtained by using an FE mesh with a discretized air gap. It then applies the model to determine the open-circuit electromotive force of an axial-flux permanent-magnet machine and compares the results with measurements.

A general design method for electric machines using magnetic circuit model considering the flux saturation problem

This paper proposes an analytical approach that is generalized for the design of various types of electric machines based on a physical magnetic circuit model. Conventional approaches have been used to predict the behavior of electric machines but have limitations in accurate flux saturation analysis and hence machine dimensioning at the initial design stage. In particular, magnetic saturation is generally ignored or compensated by correction factors in simplified models since it is difficult to determine the flux in each stator tooth for machines with any slot-pole combinations. In this paper, the flux produced by stator winding currents can be calculated accurately and rapidly for each stator tooth using the developed model, taking saturation into account. This aids machine dimensioning without the need for a computationally expensive finite element analysis (FEA). A 48-slot machine operated in induction and doubly-fed modes is used to demonstrate the proposed model. FEA is employed for verification.