MapReduce is a programming model and an associated implementation for processing and generating large data sets. Users specify a map function that processes a key/value pair to generate a set of intermediate key/value pairs, and a reduce function that merges all intermediate values associated with the same intermediate key. Many real world tasks are expressible in this model, as shown in the paper. Programs written in this functional style are automatically parallelized and executed on a large cluster of commodity machines. The run-time system takes care of the details of partitioning the input data, scheduling the program’s execution across a set of machines, handling machine failures, and managing the required inter-machine communication. This allows programmers without any experience with parallel and distributed systems to easily utilize the resources of a large distributed system. Our implementation of MapReduce runs on a large cluster of commodity machines and is highly scalable: a typical MapReduce computation processes many terabytes of data on thousands of machines. Programmers find the system easy to use: hundreds of MapReduce programs have been implemented and upwards of one thousand MapReduce jobs are executed on Google’s clusters every day. 1

A vast body of theoretical research has focused either on overly simplistic models of parallel computation, notably the PRAM, or overly specific models that have few representatives in the real world. Both kinds of models encourage exploitation of formal loopholes, rather than rewarding development of techniques that yield performance across a range of current and future parallel machines. This paper offers a new parallel machine model, called LogP, that reflects the critical technology trends underlying parallel computers. It is intended to serve as a basis for developing fast, portable parallel algorithms and to offer guidelines to machine designers. Such a model must strike a balance between detail and simplicity in order to reveal important bottlenecks without making analysis of interesting problems intractable. The model is based on four parameters that specify abstractly the computing bandwidth, the communication bandwidth, the communication delay, and the efficiency of coupling communication and computation. Portable parallel algorithms typically adapt to the machine configuration, in terms of these parameters. The utility of the model is demonstrated through examples that are implemented on the CM-5.

We present a new model of parallel computation---the LogGP model---and use it to analyze a number of algorithms, most notably, the single node scatter (one-to-all personalized broadcast). The LogGP model is an extension of the LogP model for parallel computation [CKP + 93] which abstracts the communication of fixed-sized short messages through the use of four parameters: the communication latency (L), overhead (o), bandwidth (g), and the number of processors (P ). As evidenced by experimental data, the LogP model can accurately predict communication performance when only short messages are sent (as on the CM-5) [CKP + 93, CDMS94]. However, many existing parallel machines have special support for long messages and achieve a much higher bandwidth for long messages compared to short messages (e.g., IBM SP-2, Paragon, Meiko CS-2, Ncube/2). We extend the basic LogP model with a linear model for long messages. This combination, which we call the LogGP model of parallel computation, has o...

This paper gives an overview of the implementation of Nesl, a portable nested data-parallel language. This language and its implementation are the first to fully support nested data structures as well as nested dataparallel function calls. These features allow the concise description of parallel algorithms on irregular data, such as sparse matrices and graphs. In addition, they maintain the advantages of data-parallel languages: a simple programming model and portability. The current Nesl implementation is based on an intermediate language called Vcode and a library of vector routines called Cvl. It runs on the Connection Machine CM-2, the Cray Y-MP C90, and serial machines. We compare initial benchmark results of Nesl with those of machine-specific code on these machines for three algorithms: least-squares line-fitting, median finding, and a sparse-matrix vector product. These results show that Nesl's performance is competitive with that of machine-specific codes for regular dense da...

The MIT J-Machine multicomputer has been constructed to study the role of a set of primitive mechanisms in providing efficient support for parallel computing. Each J-Machine node consists of an integrated multicomputer component, the Message-Driven Processor (MDP), and 1 MByte of DRAM. The MDP provides mechanisms to support efficient communication, synchronization, and naming. A 512 node J-Machine is operational and is due to be expanded to 1024 nodes in March 1993. In this paper we discuss the design of the J-Machine and evaluate the effectiveness of the mechanisms incorporated into the MDP. We measure the performance of the communication and synchronization mechanisms directly and investigate the behavior of four complete applications. 1 Introduction Over the past 40 years, sequential von Neumann processors have evolved a set of mechanisms appropriate for supporting most sequential programming models. It is clear, however, from efforts to build concurrent machines by connecting man...

This report describes Nesl, a strongly-typed, applicative, data-parallel language. Nesl is intended to be used as a portable interface for programming a variety of parallel and vector supercomputers, and as a basis for teaching parallel algorithms. Parallelism is supplied through a simple set of data-parallel constructs based on vectors, including a mechanism for applying any function over the elements of a vector in parallel, and a broad set of parallel functions that manipulate vectors. Nesl fully supports nested vectors and nested parallelism---the ability to take a parallel function and then apply it over multiple instances in parallel. Nested parallelism is important for implementing algorithms with complex and dynamically changing data structures, such as required in many graph or sparse matrix algorithms. Nesl also provides a mechanism for calculating the asymptotic running time for a program on various parallel machine models, including the parallel random access machine (PRAM...

We survey parallel programming models and languages using 6 criteria [:] should be easy to program, have a software development methodology, be architecture-independent, be easy to understand, guranatee performance, and provide info about the cost of programs. ... We consider programming models in 6 categories, depending on the level of abstraction they provide.

The views and conclusions contained in this document are those of the authors and should not be interpreted as necessarily representing the official policies or endorsements, either expressed or implied, of Wright Laboratory or the U. S. Government. Keywords: Data-parallel, parallel algorithms, supercomputers, nested parallelism, This report describes Nesl, a strongly-typed, applicative, data-parallel language. Nesl is intended to be used as a portable interface for programming a variety of parallel and vector computers, and as a basis for teaching parallel algorithms. Parallelism is supplied through a simple set of data-parallel constructs based on sequences, including a mechanism for applying any function over the elements of a sequence in parallel and a rich set of parallel functions that manipulate sequences. Nesl fully supports nested sequences and nested parallelism—the ability to take a parallel function and apply it over multiple instances in parallel. Nested parallelism is important for implementing algorithms with irregular nested loops (where the inner loop lengths depend on the outer iteration) and for divide-and-conquer algorithms. Nesl also provides a performance model for calculating the asymptotic performance of a program on

This paper evaluates the suitability of the MapReduce model for multi-core and multi-processor systems. MapReduce was created by Google for application development on data-centers with thousands of servers. It allows programmers to write functional-style code that is automatically parallelized and scheduled in a distributed system. We describe Phoenix, an implementation of MapReduce for shared-memory systems that includes a programming API and an efficient runtime system. The Phoenix runtime automatically manages thread creation, dynamic task scheduling, data partitioning, and fault tolerance across processor nodes. We study Phoenix with multi-core and symmetric multiprocessor systems and evaluate its performance potential and error recovery features. We also compare MapReduce code to code written in lower-level APIs such as P-threads. Overall, we establish that, given a careful implementation, MapReduce is a promising model for scalable performance on shared-memory systems with simple parallel code. 1

Experienced algorithm designers rely heavily on a set of building blocks and on the tools needed to put the blocks together into an algorithm. The understanding of these basic blocks and tools is therefore critical to the understanding of algorithms. Many of the blocks and tools needed for parallel