Fast and efficient automatic memory management for GPUs using compiler-assisted runtime coherence scheme

Exploiting the performance potential of GPUs requires managing the data transfers to and from them efficiently which is an error-prone and tedious task. In this paper, we develop a software coherence mechanism to fully automate all data transfers between the CPU and GPU without any assistance from the programmer. Our mechanism uses compiler analysis to identify potential stale accesses and uses a runtime to initiate transfers as necessary. This allows us to avoid redundant transfers that are exhibited by all other existing automatic memory management proposals. We integrate our automatic memory manager into the X10 compiler and runtime, and find that it not only results in smaller and simpler programs, but also eliminates redundant memory transfers. Tested on eight programs ported from the Rodinia benchmark suite it achieves (i) a 1.06x speedup over hand-tuned manual memory management, and (ii) a 1.29x speedup over another recently proposed compiler--runtime automatic memory management system. Compared to other existing runtime-only and compiler-only proposals, it also transfers 2.2x to 13.3x less data on average.

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.

Solar photovoltaic assistance for LHB rail coaches

Awareness for the need of sustainable and eco-friendly mobility has been increasing and various innovations are taking place in this regard. A study was carried out to assess the feasibility of installing solar photovoltaic (PV) modules atop train coaches. Most long-distance trains having LHB coaches do not have self-generating systems, thus making power cars mandatory to supply the required power for lighting loads. Feasibility of supplementing diesel generator sets with power from solar PV modules installed on coach rooftops has been reported in this communication. Not only is there a conservation of fuel, there is also a significant reduction in CO2 emissions. This work has shown that the area available on coach rooftops is more than sufficient to generate the required power, during sunlight hours, for the electrical loads of a non-A/C coach even during winter. All calculations were done keeping a standard route as the reference. Taking the cost of diesel to be Rs 66/litre, it was estimated that there will be annual savings of Rs 5,900,000 corresponding to 90,800 litres diesel per rake per year by implementing this scheme. The installation cost of solar modules would be recovered within 2-3 years. Implementation of this scheme would also amount to an annual reduction of 239 tonnes of CO2 emissions.

Precise Slicing in Imperative Programs via Term-Rewriting and Abstract Interpretation

We propose a new approach for producing precise constrained slices of programs in a language such as C. We build upon a previous approach for this problem, which is based on term-rewriting, which primarily targets loop-free fragments and is fully precise in this setting. We incorporate abstract interpretation into term-rewriting, using a given arbitrary abstract lattice, resulting in a novel technique for slicing loops whose precision is linked to the power of the given abstract lattice. We address pointers in a first-class manner, including when they are used within loops to traverse and update recursive data structures. Finally, we illustrate the comparative precision of our slices over those of previous approaches using representative examples.

Compiler/Runtime Framework for Dynamic Dataflow Parallelization of Tiled Programs

Task-parallel languages are increasingly popular. Many of them provide expressive mechanisms for intertask synchronization. For example, OpenMP 4.0 will integrate data-driven execution semantics derived from the StarSs research language. Compared to the more restrictive data-parallel and fork-join concurrency models, the advanced features being introduced into task-parallelmodels in turn enable improved scalability through load balancing, memory latency hiding, mitigation of the pressure on memory bandwidth, and, as a side effect, reduced power consumption. In this article, we develop a systematic approach to compile loop nests into concurrent, dynamically constructed graphs of dependent tasks. We propose a simple and effective heuristic that selects the most profitable parallelization idiom for every dependence type and communication pattern. This heuristic enables the extraction of interband parallelism (cross-barrier parallelism) in a number of numerical computations that range from linear algebra to structured grids and image processing. The proposed static analysis and code generation alleviates the burden of a full-blown dependence resolver to track the readiness of tasks at runtime. We evaluate our approach and algorithms in the PPCG compiler, targeting OpenStream, a representative dataflow task-parallel language with explicit intertask dependences and a lightweight runtime. Experimental results demonstrate the effectiveness of the approach.