Jeremy Singer

The Glasgow Raspberry Pi Cloud: A Scale Model for Cloud Computing Infrastructures

Data Centers (DC) used to support Cloud services often consist of tens of thousands of networked machines under a single roof. The significant capital outlay required to replicate such infrastructures constitutes a major obstacle to practical implementation and evaluation of research in this domain. Currently, most research into Cloud computing relies on either limited software simulation, or the use of a testbed environments with a handful of machines. The recent introduction of the Raspberry Pi, a low-cost, low-power single-board computer, has made the construction of a miniature Cloud DCs more affordable. In this paper, we present the Glasgow Raspberry Pi Cloud (PiCloud), a scale model of a DC composed of clusters of Raspberry Pi devices. The PiCloud emulates every layer of a Cloud stack, ranging from resource virtualisation to network behaviour, providing a full-featured Cloud Computing research and educational environment.

Intelligent selection of application-specific garbage collectors

Java program execution times vary greatly with different garbage collection algorithms. Until now, it has not been possible to determine the best GC algorithm for aparticular program without exhaustively profiling that program for all available GC algorithms. This paper presents a new approach. We use machine learning techniques to build a prediction model that, given asingle profile run of a previously unseen Java program,can predict a good GC algorithm for that program. We implement this technique in Jikes RVM and test it onseveral standard benchmark suites. Our techniqueachieves 5% speedup in overall execution time (averagedacross all test programs for all heap sizes) compared with selecting the default GC algorithm in every trial. We present further experiments to show that an oracle predictor could achieve an average 17% speedup on the same experiments. In addition, we provide evidence to suggest that GC behaviour is sometimes independent of program inputs. These observations lead us to propose that intelligent selection of GC algorithms is suitably straight forward, efficient and effective to merit further exploration regarding its potential inclusion in the general Java software deployment process.

Fundamental Nano-Patterns to Characterize and Classify Java Methods

Fundamental nano-patterns are simple, static, binary properties of Java methods, such as ObjectCreator and Recursive. We present a provisional catalogue of 17 such nano-patterns. We report statistical and information theoretic metrics to show the frequency of nano-pattern occurrence in a large corpus of open-source Java projects. We proceed to give two example case studies that demonstrate potential applications for nano-patterns. The first study involves a quantitative comparison of two popular Java benchmarking suites, in terms of their relative object-orientedness and diversity. The second study involves applying machine learning techniques to program comprehension, using method nano-patterns as learning features. In both studies, nano-patterns provide concise summaries of Java methods to enable efficient and effective analysis.

Garbage collection auto-tuning for Java mapreduce on multi-cores

MapReduce has been widely accepted as a simple programming pattern that can form the basis for efficient, large-scale, distributed data processing. The success of the MapReduce pattern has led to a variety of implementations for different computational scenarios. In this paper we present MRJ, a MapReduce Java framework for multi-core architectures. We evaluate its scalability on a four-core, hyperthreaded Intel Core i7 processor, using a set of standard MapReduce benchmarks. We investigate the significant impact that Java runtime garbage collection has on the performance and scalability of MRJ. We propose the use of memory management auto-tuning techniques based on machine learning. With our auto-tuning approach, we are able to achieve MRJ performance within 10% of optimal on 75% of our benchmark tests.

The economics of garbage collection

This paper argues that economic theory can improve our understanding of memory management. We introduce the allocation curve, as an analogue of the demand curve from microeconomics. An allocation curve for a program characterises how the amount of garbage collection activity required during its execution varies in relation to the heap size associated with that program. The standard treatment of microeconomic demand curves (shifts and elasticity) can be applied directly and intuitively to our new allocation curves. As an application of this new theory, we show how allocation elasticity can be used to control the heap growth rate for variable sized heaps in Jikes RVM.

Exploiting the Correspondence between Micro Patterns and Class Names

This paper argues that semantic information encoded in natural language identifiers is a largely neglected resource for program analysis. First we show that words in Java class names relate to class properties, expressed using the recently developed micro patterns language. We analyse a large corpus of Java programs to create a database that links common class name words with micro patterns. Finally we report on prototype tools integrated with the Eclipse development environment. These tools use the database to inform programmers of particular problems or optimization opportunities in their code.

Search-Based Refactoring: Metrics Are Not Enough

Search-based Software Engineering (SBSE) techniques have been applied extensively to refactor software, often based on metrics that describe the object-oriented structure of an application. Recent work shows that in some cases applying popular SBSE tools to open-source software does not necessarily lead to an improved version of the software as assessed by some subjective criteria. Through a survey of professionals, we investigate the relationship between popular SBSE refactoring metrics and the subjective opinions of software engineers. We find little or no correlation between the two. Through qualitative analysis, we find that a simple static view of software is insufficient to assess software quality, and that software quality is dependent on factors that are not amenable to measurement via metrics. We recommend that future SBSE refactoring research should incorporate information about the dynamic behaviour of software, and conclude that a human-in-the-loop approach may be the only way to refactor software in a manner helpful to an engineer.

Toward a more accurate understanding of the limits of the TLS execution paradigm

Thread-Level Speculation (TLS) facilitates the extraction of parallel threads from sequential applications. Most prior work has focused on developing the compiler and architecture for this execution paradigm. Such studies often narrowly concentrated on a specific design point. On the other hand, other studies have attempted to assess how well TLS performs if some architectural/ compiler constraint is relaxed. Unfortunately, such previous studies have failed to truly assess TLS performance potential, because they have been bound to some specific TLS architecture and have ignored one or another important TLS design choice, such as support for out-of-order task spawn or support for intermediate checkpointing.

Online non-stationary boosting

Ozas Online Boosting algorithm provides a version of AdaBoost which can be trained in an online way for stationary problems. One perspective is that this enables the power of the boosting framework to be applied to datasets which are too large to fit into memory. The online boosting algorithm assumes the data distribution to be independent and identically distributed (i.i.d.) and therefore has no provision for concept drift. We present an algorithm called Online Non-Stationary Boosting (ONSBoost) that, like Online Boosting, uses a static ensemble size without generating new members each time new examples are presented, and also adapts to a changing data distribution. We evaluate the new algorithm against Online Boosting, using the STAGGER dataset and three challenging datasets derived from a learning problem inside a parallelising virtual machine. We find that the new algorithm provides equivalent performance on the STAGGER dataset and an improvement of up to 3% on the parallelisation datasets.

JVM-hosted languages: they talk the talk, but do they walk the walk?

The rapid adoption of non-Java JVM languages is impressive: major international corporations are staking critical parts of their software infrastructure on components built from languages such as Scala and Clojure. However with the possible exception of Scala, there has been little academic consideration and characterization of these languages to date. In this paper, we examine four non-Java JVM languages and use exploratory data analysis techniques to investigate differences in their dynamic behavior compared to Java. We analyse a variety of programs and levels of behavior to draw distinctions between the different programming languages. We briefly discuss the implications of our findings for improving the performance of JIT compilation and garbage collection on the JVM platform.