Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userArchitecture-Specific Performance Optimization of Compute-Intensive FaaS Functions
74
0
0.0
(
0
)
Added by
Mohak Chadha
Publication date
2021
fields
Informatics Engineering
and research's language is
English
Ask ChatGPT about the research
No Arabic abstract
FaaS allows an application to be decomposed into functions that are executed on a FaaS platform. The FaaS platform is responsible for the resource provisioning of the functions. Recently, there is a growing trend towards the execution of compute-intensive FaaS functions that run for several seconds. However, due to the billing policies followed by commercial FaaS offerings, the execution of these functions can incur significantly higher costs. Moreover, due to the abstraction of underlying processor architectures on which the functions are executed, the performance optimization of these functions is challenging. As a result, most FaaS functions use pre-compiled libraries generic to x86-64 leading to performance degradation. In this paper, we examine the underlying processor architectures for Google Cloud Functions (GCF) and determine their prevalence across the 19 available GCF regions. We modify, adapt, and optimize three compute-intensive FaaS workloads written in Python using Numba, a JIT compiler based on LLVM, and present results wrt performance, memory consumption, and costs on GCF. Results from our experiments show that the optimization of FaaS functions can improve performance by 12.8x (geometric mean) and save costs by 73.4% on average for the three functions. Our results show that optimization of the FaaS functions for the specific architecture is very important. We achieved a maximum speedup of 1.79x by tuning the function especially for the instruction set of the underlying processor architecture.
rate research
Read More
Data-intensive applications impact many domains, and their steadily increasing size and complexity demands high-performance, highly usable environments. We integrate a set of ideas developed in various data science and data engineering frameworks. They employ a set of operators on specific data abstractions that include vectors, matrices, tensors, graphs, and tables. Our key concepts are inspired from systems like MPI, HPF (High-Performance Fortran), NumPy, Pandas, Spark, Modin, PyTorch, TensorFlow, RAPIDS(NVIDIA), and OneAPI (Intel). Further, it is crucial to support different languages in everyday use in the Big Data arena, including Python, R, C++, and Java. We note the importance of Apache Arrow and Parquet for enabling language agnostic high performance and interoperability. In this paper, we propose High-Performance Tensors, Matrices and Tables (HPTMT), an operator-based architecture for data-intensive applications, and identify the fundamental principles needed for performance and usability success. We illustrate these principles by a discussion of examples using our software environments, Cylon and Twister2 that embody HPTMT.
This paper investigates the multi-GPU performance of a 3D buoyancy driven cavity solver using MPI and OpenACC directives on different platforms. The paper shows that decomposing the total problem in different dimensions affects the strong scaling performance significantly for the GPU. Without proper performance optimizations, it is shown that 1D domain decomposition scales poorly on multiple GPUs due to the noncontiguous memory access. The performance using whatever decompositions can be benefited from a series of performance optimizations in the paper. Since the buoyancy driven cavity code is latency-bounded on the clusters examined, a series of optimizations both agnostic and tailored to the platforms are designed to reduce the latency cost and improve memory throughput between hosts and devices efficiently. First, the parallel message packing/unpacking strategy developed for noncontiguous data movement between hosts and devices improves the overall performance by about a factor of 2. Second, transferring different data based on the stencil sizes for different variables further reduces the communication overhead. These two optimizations are general enough to be beneficial to stencil computations having ghost changes on all of the clusters tested. Third, GPUDirect is used to improve the communication on clusters which have the hardware and software support for direct communication between GPUs without staging CPUs memory. Finally, overlapping the communication and computations is shown to be not efficient on multi-GPUs if only using MPI or MPI+OpenACC. Although we believe our implementation has revealed enough overlap, the actual running does not utilize the overlap well due to a lack of asynchronous progression.
The rigid MPI programming model and batch scheduling dominate high-performance computing. While clouds brought new levels of elasticity into the world of computing, supercomputers still suffer from low resource utilization rates. To enhance supercomputing clusters with the benefits of serverless computing, a modern cloud programming paradigm for pay-as-you-go execution of stateless functions, we present rFaaS, the first RDMA-aware Function-as-a-Service (FaaS) platform. With hot invocations and decentralized function placement, we overcome the major performance limitations of FaaS systems and provide low-latency remote invocations in multi-tenant environments. We evaluate the new serverless system through a series of microbenchmarks and show that remote functions execute with negligible performance overheads. We demonstrate how serverless computing can bring elastic resource management into MPI-based high-performance applications. Overall, our results show that MPI applications can benefit from modern cloud programming paradigms to guarantee high performance at lower resource costs.
A novel derivative-free algorithm, optimization by moving ridge functions (OMoRF), for unconstrained and bound-constrained optimization is presented. This algorithm couples trust region methodologies with output-based dimension reduction to accelerate convergence of model-based optimization strategies. The dimension-reducing subspace is updated as the trust region moves through the function domain, allowing OMoRF to be applied to functions with no known global low-dimensional structure. Furthermore, its low computational requirement allows it to make rapid progress when optimizing high-dimensional functions. Its performance is examined on a set of test problems of moderate to high dimension and a high-dimensional design optimization problem. The results show that OMoRF compares favourably to other common derivative-free optimization methods, even for functions in which no underlying global low-dimensional structure is known.
Big data applications and analytics are employed in many sectors for a variety of goals: improving customers satisfaction, predicting market behavior or improving processes in public health. These applications consist of complex software stacks that are often run on cloud systems. Predicting execution times is important for estimating the cost of cloud services and for effectively managing the underlying resources at runtime. Machine Learning (ML), providing black box solutions to model the relationship between application performance and system configuration without requiring in-detail knowledge of the system, has become a popular way of predicting the performance of big data applications. We investigate the cost-benefits of using supervised ML models for predicting the performance of applications on Spark, one of todays most widely used frameworks for big data analysis. We compare our approach with textit{Ernest} (an ML-based technique proposed in the literature by the Spark inventors) on a range of scenarios, application workloads, and cloud system configurations. Our experiments show that Ernest can accurately estimate the performance of very regular applications, but it fails when applications exhibit more irregular patterns and/or when extrapolating on bigger data set sizes. Results show that our models match or exceed Ernests performance, sometimes enabling us to reduce the prediction error from 126-187% to only 5-19%.
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
