A Significance-Driven Programming Framework for Energy-Constrained Approximate Computing

Approximate execution is a viable technique for energy-con\-strained environments, provided that applications have the mechanisms to produce outputs of the highest possible quality within the given energy budget.
We introduce a framework for energy-constrained execution with controlled and graceful quality loss. A simple programming model allows users to express the relative importance of computations for the quality of the end result, as well as minimum quality requirements. The significance-aware runtime system uses an application-specific analytical energy model to identify the degree of concurrency and approximation that maximizes quality while meeting user-specified energy constraints. Evaluation on a dual-socket 8-core server shows that the proposed
framework predicts the optimal configuration with high accuracy, enabling energy-constrained executions that result in significantly higher quality compared to loop perforation, a compiler approximation technique.

On the potential of significance-driven execution for energy-aware HPC

Dynamic Voltage and Frequency Scaling (DVFS) exhibits fundamental limitations as a method to reduce energy consumption in computing systems. In the HPC domain, where performance is of highest priority and codes are heavily optimized to minimize idle time, DVFS has limited opportunity to achieve substantial energy savings. This paper explores if operating processors Near the transistor Threshold Volt- age (NTV) is a better alternative to DVFS for break- ing the power wall in HPC. NTV presents challenges, since it compromises both performance and reliability to reduce power consumption. We present a first of its kind study of a significance-driven execution paradigm that selectively uses NTV and algorithmic error tolerance to reduce energy consumption in performance- constrained HPC environments. Using an iterative algorithm as a use case, we present an adaptive execution scheme that switches between near-threshold execution on many cores and above-threshold execution on one core, as the computational significance of iterations in the algorithm evolves over time. Using this scheme on state-of-the-art hardware, we demonstrate energy savings ranging between 35% to 67%, while compromising neither correctness nor performance.

Significance Driven Computation: A Voltage-Scalable, Variation-Aware, Quality-Tuning Motion Estimator

In this paper we present a design methodology for algorithm/architecture co-design of a voltage-scalable, process variation aware motion estimator based on significance driven computation. The fundamental premise of our approach lies in the fact that all computations are not equally significant in shaping the output response of video systems. We use a statistical technique to intelligently identify these significant/not-so-significant computations at the algorithmic level and subsequently change the underlying architecture such that the significant computations are computed in an error free manner under voltage over-scaling. Furthermore, our design includes an adaptive quality compensation (AQC) block which "tunes" the algorithm and architecture depending on the magnitude of voltage over-scaling and severity of process variations. Simulation results show average power savings of similar to 33% for the proposed architecture when compared to conventional implementation in the 90 nm CMOS technology. The maximum output quality loss in terms of Peak Signal to Noise Ratio (PSNR) was similar to 1 dB without incurring any throughput penalty.

Significance Driven Computation on Next-Generation Unreliable Platforms

In this paper, we propose a design paradigm for energy efficient and variation-aware operation of next-generation multicore heterogeneous platforms. The main idea behind the proposed approach lies on the observation that not all operations are equally important in shaping the output quality of various applications and of the overall system. Based on such an observation, we suggest that all levels of the software design stack, including the programming model, compiler, operating system (OS) and run-time system should identify the critical tasks and ensure correct operation of such tasks by assigning them to dynamically adjusted reliable cores/units. Specifically, based on error rates and operating conditions identified by a sense-and-adapt (SeA) unit, the OS selects and sets the right mode of operation of the overall system. The run-time system identifies the critical/less-critical tasks based on special directives and schedules them to the appropriate units that are dynamically adjusted for highly-accurate/approximate operation by tuning their voltage/frequency. Units that execute less significant operations can operate at voltages less than what is required for correct operation and consume less power, if required, since such tasks do not need to be always exact as opposed to the critical ones. Such scheme can lead to energy efficient and reliable operation, while reducing the design cost and overheads of conventional circuit/micro-architecture level techniques.

Energy Efficiency through Significance-Based Computing

An extension of approximate computing, significance-based computing exploits applications' inherent error resiliency and offers a new structural paradigm that strategically relaxes full computational precision to provide significant energy savings with minimal performance degradation.

A novel energy-driven computing paradigm for e-health scenarios

A first-rate e-Health system saves lives, provides better patient care, allows complex but useful epidemiologic analysis and saves money. However, there may also be concerns about the costs and complexities associated with e-health implementation, and the need to solve issues about the energy footprint of the high-demanding computing facilities. This paper proposes a novel and evolved computing paradigm that: (i) provides the required computing and sensing resources; (ii) allows the population-wide diffusion; (iii) exploits the storage, communication and computing services provided by the Cloud; (iv) tackles the energy-optimization issue as a first-class requirement, taking it into account during the whole development cycle. The novel computing concept and the multi-layer top-down energy-optimization methodology obtain promising results in a realistic scenario for cardiovascular tracking and analysis, making the Home Assisted Living a reality.

Mitigating the Impact of Faults in Unreliable Memories for Error-Resilient Applications

Inherently error-resilient applications in areas such as signal processing, machine learning and data analytics provide opportunities for relaxing reliability requirements, and thereby reducing the overhead incurred by conventional error correction schemes. In this paper, we exploit the tolerable imprecision of such applications by designing an energy-efficient fault-mitigation scheme for unreliable data memories to meet target yield. The proposed approach uses a bit-shuffling mechanism to isolate faults into bit locations with lower significance. This skews the bit-error distribution towards the low order bits, substantially limiting the output error magnitude. By controlling the granularity of the shuffling, the proposed technique enables trading-off quality for power, area, and timing overhead. Compared to error-correction codes, this can reduce the overhead by as much as 83% in read power, 77% in read access time, and 89% in area, when applied to various data mining applications in 28nm process technology.

Specification-driven approach for protocol design of distributed computing systems

An important issue in the design of a distributed computing system (DCS) is the development of a suitable protocol. This paper presents an effort to systematize the protocol design procedure for a DCS. Protocol design and development can be divided into six phases: specification of the DCS, specification of protocol requirements, protocol design, specification and validation of the designed protocol, performance evaluation, and hardware/software implementation. This paper describes techniques for the second and third phases, while the first phase has been considered by the authors in their earlier work. Matrix and set theoretic based approaches are used for specification of a DCS and for specification of the protocol requirements. These two formal specification techniques form the basis of the development of a simple and straightforward procedure for the design of the protocol. The applicability of the above design procedure has been illustrated by considering an example of a computing system encountered on board a spacecraft. A Petri-net based approach has been adopted to model the protocol. The methodology developed in this paper can be used in other DCS applications.

A Programming Model and Runtime System for Significance-Aware Energy-Efficient Computing.

We introduce a task-based programming model and runtime system that exploit the observation that not all parts of a program are equally significant for the accuracy of the end-result, in order to trade off the quality of program outputs for increased energy-efficiency. This is done in a structured and flexible way, allowing for easy exploitation of different points in the quality/energy space, without adversely affecting application performance. The runtime system can apply a number of different policies to decide whether it will execute less-significant tasks accurately or approximately.

The experimental evaluation indicates that our system can achieve an energy reduction of up to 83% compared with a fully accurate execution and up to 35% compared with an approximate version employing loop perforation. At the same time, our approach always results in graceful quality degradation.

Towards Automated Data-Driven Model Creation for Cloud Computing Simulation

The increasing complexity and scale of cloud computing environments due to widespread data centre heterogeneity makes measurement-based evaluations highly difficult to achieve. Therefore the use of simulation tools to support decision making in cloud computing environments to cope with this problem is an increasing trend. However the data required in order to model cloud computing environments with an appropriate degree of accuracy is typically large, very difficult to collect without some form of automation, often not available in a suitable format and a time consuming process if done manually. In this research, an automated method for cloud computing topology definition, data collection and model creation activities is presented, within the context of a suite of tools that have been developed and integrated to support these activities.

Exploiting Significance of Computations for Energy-Constrained Approximate Computing

Approximate execution is a viable technique for environments with energy constraints, provided that applications are given the mechanisms to produce outputs of the highest possible quality within the available energy budget. This paper introduces a framework for energy-constrained execution with controlled and graceful quality loss. A simple programming model allows developers to structure the computation in different tasks, and to express the relative importance of these tasks for the quality of the end result. For non-significant tasks, the developer can also supply less costly, approximate versions. The target energy consumption for a given execution is specified when the application is launched. A significance-aware runtime system employs an application-specific analytical energy model to decide how many cores to use for the execution, the operating frequency for these cores, as well as the degree of task approximation, so as to maximize the quality of the output while meeting the user-specified energy constraints. Evaluation on a dual-socket 16-core Intel platform using 9 benchmark kernels shows that the proposed framework picks the optimal configuration with high accuracy. Also, a comparison with loop perforation (a well-known compile-time approximation technique), shows that the proposed framework results in significantly higher quality for the same energy budget.

Improving Management of Distributed Services Using Correlations and Predictions in SLA-Driven Cloud Computing Systems

Recent advancements in cloud computing have enabled the proliferation of distributed applications, which require management and control of multiple services. However, without an efficient mechanism for scaling services in response to changing environmental conditions and number of users, application performance might suffer, leading to Service Level Agreement (SLA) violations and inefficient use of hardware resources. We introduce a system for controlling the complexity of scaling applications composed of multiple services using mechanisms based on fulfillment of SLAs. We present how service monitoring information can be used in conjunction with service level objectives, predictions, and correlations between performance indicators for optimizing the allocation of services belonging to distributed applications. We validate our models using experiments and simulations involving a distributed enterprise information system. We show how discovering correlations between application performance indicators can be used as a basis for creating refined service level objectives, which can then be used for scaling the application and improving the overall application's performance under similar conditions.

Service Level Agreements-Driven Management of Distributed Applications in Cloud Computing Environments

Cloud Computing enables provisioning and distribution of highly scalable services in a reliable, on-demand and sustainable manner. However, objectives of managing enterprise distributed applications in cloud environments under Service Level Agreement (SLA) constraints lead to challenges for maintaining optimal resource control. Furthermore, conflicting objectives in management of cloud infrastructure and distributed applications might lead to violations of SLAs and inefficient use of hardware and software resources. This dissertation focusses on how SLAs can be used as an input to the cloud management system, increasing the efficiency of allocating resources, as well as that of infrastructure scaling. First, we present an extended SLA semantic model for modelling complex service-dependencies in distributed applications, and for enabling automated cloud infrastructure management operations. Second, we describe a multi-objective VM allocation algorithm for optimised resource allocation in infrastructure clouds. Third, we describe a method of discovering relations between the performance indicators of services belonging to distributed applications and then using these relations for building scaling rules that a CMS can use for automated management of VMs. Fourth, we introduce two novel VM-scaling algorithms, which optimally scale systems composed of VMs, based on given SLA performance constraints. All presented research works were implemented and tested using enterprise distributed applications.