CPU Power Research Articles

Modeling and simulation of power consumption and execution times for real-time tasks on embedded heterogeneous architectures

In this work, we introduce a power-consumption model for heterogeneous multicore architectures that captures the variability of energy consumption based on processing workload type, in addition to the classical variables considered in the literature, like type and frequency of the CPU. We motivate the approach presenting experimental results gathered on a Odroid-XU3 board equipped with an Arm big.LITTLE CPU, showing that power consumption has a non-negligible dependency on the workload type. We also present a model to define the execution time of the tasks, which depends on both the workload, and the CPU frequency and architecture. We present our modifications to the open-source RTSIM real-time scheduling simulator to extend its CPU power consumption and execution time duration models, integrating results taken from the real platform. The presented work constitutes a useful base for future research in power-aware real-time scheduling on heterogeneous platforms.

TrCMP: A dependable app usage inference design for user behavior analysis through cyber-physical parameters

Abstract Timely and accurate service recommendations can help users improve their quality of life and efficiency of work. The accuracy of service recommendations often depends on effective user behavior analysis. Technically, user behaviors associated with a specific service could be reflected in the apps they use, while operating different apps will change cyber-physical parameters on Android devices, such as power consumption, CPU usage, memory usage, and traffic data. Some studies have already leveraged cyber-physical parameters to analyze app usage. However, these investigations usually take into consideration a single parameter (e.g., traffic or power), which could be easily affected by the complex Android environment. Moreover, to the best of our knowledge, these investigations cannot analyze some apps that run in the offline state. In this paper, we study the similarities and the differences of parameters with various apps running in both online and offline states. Then we propose a design of a dependable app usage inference, named TrCMP to understand the inference in a mobile system through a combination of the cyber and physical system parameters. This design comprehensively considers Tr affic, C PU, M emory and P ower to speculate apps running in both online and offline states. An algorithm is proposed to find the most effective weight values for each parameter. Extensive experiments are conducted to validate the effectiveness and dependability of our design.

Impacted Cyber Attacks Assessment in Wide Range of Big Data Security Systems

The technological advancements in image storage, data processing, and signal analysis of Big Data include (a) the fastly degrade the cost of storage and CPU power in recent arena; the flexibility and cost-effectiveness of data operating platforms and cloud computing systems for flexible computation and storage; and (c) the development of new frameworks , which allow users to take advantage of these divided computing systems storing large amount of data which is almost flexible parallel processing. The proposed survey work focused on discussing the various impacted cyber-attack critics available in industry and the trending algorithms available for cyber security etc. Big data in IoT clouds handling and software platforms which allow the malware enter into the working systems are analyzed, reliable methods to avoid the miscellaneous malwares are clearly depicted here.

Optimal Task Offloading and Resource Allocation in Mobile-Edge Computing With Inter-User Task Dependency

Mobile-edge computing (MEC) has recently emerged as a cost-effective paradigm to enhance the computing capability of hardware-constrained wireless devices (WDs). In this paper, we first consider a two-user MEC network, where each WD has a sequence of tasks to execute. In particular, we consider task dependency between the two WDs, where the input of a task at one WD requires the final task output at the other WD. Under the considered task-dependency model, we study the optimal task offloading policy and resource allocation (e.g., on offloading transmit power and local CPU frequencies) that minimize the weighted sum of the WDs' energy consumption and task execution time. The problem is challenging due to the combinatorial nature of the offloading decisions among all tasks and the strong coupling with resource allocation. To tackle this problem, we first assume that the offloading decisions are given and derive the closed-form expressions of the optimal offloading transmit power and local CPU frequencies. Then, an efficient bi-section search method is proposed to obtain the optimal solutions. Furthermore, we prove that the optimal offloading decisions follow an one-climb policy, based on which a reduced-complexity Gibbs Sampling algorithm is proposed to obtain the optimal offloading decisions. We then extend the investigation to a general multi-user scenario, where the input of a task at one WD requires the final task outputs from multiple other WDs. Numerical results show that the proposed method can significantly outperform the other representative benchmarks and efficiently achieve low complexity with respect to the call graph size.

Physical Unclonable Function-Based Key Sharing via Machine Learning for IoT Security

In many industry Internet of Things applications, resources like CPU, memory, and battery power are limited and cannot afford the classic cryptographic security solutions. Silicon physical unclonable function (PUF) is a lightweight security primitive that exploits manufacturing variations during the chip fabrication process for key generation and/or device authentication. However, traditional weak PUFs such as ring oscillator (RO) PUF generate chip-unique key for each device, which restricts their application in security protocols where the same key is required to be shared in resource-constrained devices. In this article, in order to address this issue, we propose a PUF-based key sharing method for the first time. The basic idea is to implement one-to-one input–output mapping with lookup table (LUT)-based interstage crossing structures in each level of inverters of RO PUF. Individual customization on configuration bits of interstage crossing structure and different RO selections with challenges bring high flexibility. Therefore, with the flexible configuration of interstage crossing structures and challenges, crossover RO PUF can generate the same shared key for resource-constrained devices, which enables a new application for lightweight key sharing protocols.

Reducing CPU Power Consumption with Device Utilization-Aware DVFS for Low-Latency SSDs

Reducing CPU Power Consumption with Device Utilization-Aware DVFS for Low-Latency SSDs

Research on Various Cryptography Techniques

Security plays a critical position in preserving information privacy and secrecy. Many encryption strategies are available to protect data during transmission or storage. These encryption methods vary in terms of strength, speed, and resource consumption (CPU usage, memory, and power). This study aims to present the most popular and interesting algorithms currently in use

Advanced Efficient Energy Method Query Processing

Web search engine are made up by countless inquiry handling nodes, i.e., web servers devoted to refine individual inquiries. Such numerous web servers eat a considerable quantity of power, mainly answerable to their CPUs, however they are essential to make certain reduced latencies, because customers anticipate sub-second action times. Nevertheless, customers can rarely discover action times that are much faster than their assumptions. For this reason, we suggest the Predictive Power Conserving Online Organizing Formula to pick one of the most suitable CPU regularity to refine a question on a per-core basis. The proposed algorithm target at procedure questions by their due dates, as well as utilize top-level organizing details to minimize the CPU power usage of an inquiry handling node. PESOS base its choice on inquiry effectiveness forecasters, approximating the handling quantity and also processing time of question. End results disclose that chosen algorithm can reduce the CPU power use of a concern taking care of node roughly ~ 48% contrasted to a system doing at maximum CPU core uniformity. The proposed technique exceeds furthermore the best sophisticated competitor with a ~ 20% power preserving, while the competing ask for a wonderful spec adjusting in addition to it could maintain in irrepressible latency offenses.

Energy Consumption Between Routing and Bridging Protocol for Eco-Friendly Wireless Networks

The implementation of eco-friendly technology has been become an interesting field for sustainability. No exception with the implementation of wireless technology that used for developing networks infrastructure, it is necessary for saving the usage of energy. As a data forwarding protocol in a computer network, commonly there are two protocols that used for, which are routing and bridging protocol. Technically routing protocol has been confirmed that it is more effective and efficient than bridging protocol. However bridging protocol still becomes the popular protocol for data forwarding because it is easy to use. This research tried to test the energy consumption of the wireless network device that implementing between routing or bridging protocol. The wireless network device that used for this research was MikroTik router RB 433Ah. The data forwarding protocol that was tested consists of bridging, static routing, and RIP routing. Data traffic scenario that used for this research consisted of two scenarios which were HTML data access with packet size 256B and video streaming data access with packet size 1518B. Measuring the energy consumption referred to three parameters which were power consumption, CPU usage, and processor temperature. The result showed that for HTML data access scenario, the RIP routing protocol become the lowest energy consumption with power consumption reached 7.460 W, CPU usage 4.6 %, and processor temperature 38.133^C. While for video streaming scenario, generally the RIP routing protocol still become the lowest energy consumption with power consumption reached 7.567 W, CPU usage 7.33 %, and processor temperature 36.727^C.

DimmStore

Memory can consume a substantial amount of power in database servers, yet memory power has received considerably less attention than CPU power. Memory power consumption is also highly non-proportional. Thus, memory power becomes even more significant in the common case in which a database server is either not completely busy or not completely full. In this paper, we study the application of two memory power optimization techniques - rank-aware allocation and rate-based layout - to database systems. By concentrating memory load, rather than spreading it out evenly, these techniques create and exploit memory idleness to achieve power savings. We have implemented these techniques in a prototype database system called DimmStore. DimmStore is part of a memory power testbed which includes customized hardware with direct power measurement capabilities, allowing us to measure the techniques' effectiveness. We use the testbed to empirically characterize the power saving opportunities provided by these techniques, as well as their performance impact, under YCSB and TPC-C workloads. Under simple YCSB workloads, power savings ranged up to 50%, depending on load and space utilization, with little performance impact. Savings were smaller, but still significant, for TPC-C, which has more complex data locality characteristics.

Offload Computation in Mobile Edge Computing with Wide-Band Support of Channel Bonding

Extensive enhancement in technology enables the use of mobile and Internet of Things (IoT) devices in the high demanding applications and services. The users' demands for quality of service (QoS) and computation intensive applications such as augmented reality, online gaming, e-health, smart home and environmental monitoring etc. are also increasing exponentially [1]. Although, mobile and IoT devices have become much more powerful in the last decades and can perform a variety of tasks in an efficient manner, these are still incapable for computation intensive and time critical applications [2]. With the low processing capacity, IoT devices cannot produce results within a given time constraint. Even with high CPU power, executing a computational expensive task depletes their battery energy and shortens the operational lifetime. Mobile cloud computing (MCC) is one of the most promising solutions in the last decade [3]. It empowers the resource constrained mobile and IoT devices with elastic computing power and large storage capacity. MCC enables the realization of centralized computing and allows offload computation for CPU hungry and time sensitive tasks at the cloud server. However, intrinsically the MCC has certain serious concerns such as long transmission latency, privacy/security of user data, immense increase in Internet traffic due to billions of IoT devices in the near future [4]. These issues motivate the emerging a new paradigm known as mobile edge computing (MEC). MEC decouples the cloud resources into edge servers which are located near the end user, usually alongside access point (AP).

Hybrid Parallel FDTD Calculation Method Based on MPI for Electrically Large Objects

At present, the Internet of Things (IoT) has attracted more and more researchers' attention. Electromagnetic scattering calculation usually has the characteristics of large-scale calculation, high space-time complexity, and high precision requirement. For the background and objectives of complex environment, it is difficult for a single computer to achieve large-scale electromagnetic scattering calculation and to obtain corresponding large data. Therefore, we use Finite-Difference Time-Domain (FDTD) combined with Internet of Things, cloud computing, and other technologies to solve the above problems. In this paper, we focus on the FDTD method and use it to simulate electromagnetic scattering of electrically large objects. FDTD method has natural parallelism. A computing network cluster based on MPI is constructed. POSIX (Portable Operating System Interface of UNIX) multithreading technology is conducive to enhancing the computing power of multicore CPU and to realize multiprocessor multithreading hybrid parallel FDTD. For two-dimension CPU and memory resources, the Dominant Resource Fairness (DRF) algorithm is used to achieve load balancing scheduling, which guarantees the computing performance. The experimental results show that the hybrid parallel FDTD algorithm combined with load balancing scheduling can solve the problem of low computational efficiency and improve the success rate of task execution.

In brief

PAGE 711 Researchers from China provide an experimental study of recovery time in a bulk gallium arsenide (GaAs) avalanche semiconductor switch. The switch, operating in a low energy-triggered mode, was found to have a recovery time reduction when its biased voltage was reduced. PAGE 705 Phillips Photonics researchers from both Germany and the Netherlands have produced a compact optically pumped vertical external-cavity surface-emitting laser (OP-VECSEL) which, consisting of an optically pumped semiconductor switch and surrounded by VCSEL chips mounted on a single heat sink, allows it to reach 10 Watts of power. External view of bulk avalanche semiconductor. PAGE 688 Conventional mobile devices have thermal management problems primarily managed with software-based techniques due to the need for compact size inhibited hardware solutions. Since software-based solutions inevitably inhibit device performance a team from Korea has developed a control algorithm which adaptively controls threshold performance based on CPU power efficiency in real time. Drawing scheme of VECSEL with compact design. PAGE 701 A pair of researchers from China present a multi-order bandpass filter based on unit cells. The fabricated filter features an ultra-wide stopband and impressive stopband rejection performance as confirmed by experimental results. The easily-tunable BPF topology is suitable for a variety of applications in microwave communication systems. Power consumption measurements. PAGE 707 An active clamping circuit for voltage-sharing among series-connected insulated gate bipolar translators during the turning-on process has been presented by a team of researchers in China. The design includes a time detection circuit to detect the pulse signal indicating the working time of the active clamping circuit for driving signal regulation. Bandpass filter unit cell. Proposed active clamping circuit for turning-on process.

Design and analysis of behaviour based DDoS detection algorithm for data centres in cloud

In the recent years, security is the major concern in cloud computing. One of the major security threats to the cloud is distributed denial of service (DDoS). This DDoS attacks results in data corruption, losing of sensitive information and leads to denial of cloud service. To overcome the drawbacks of security, we proposed the behaviour-based DDoS detection algorithm that is based on the behaviour of the user which generates the traffic. The proposed work is divided into two phases which can be executed in parallel. In general, the hackers or illegitimate user floods the traffic and uses more bandwidth. In first phase, the analysis of the dynamic traffic and effectively differentiating the genuine traffic and attacker’s traffic is done in the proposed algorithm. The packet analyzer is responsible to identify the genuine packets from the flooding traffic. In second phase, the CPU utilization is monitored and the process consuming more CPU power or the source is illegitimate, the process is rejected. The simulation is carried by using the OPNET simulator. The simulation results are tested with three different scenarios. It is proved that the proposed model has efficient response time and it has better efficiency in DDoS prone zones.

Selective Mobile Cloud Offloading to Augment Multi-Persona Performance and Viability

Fueled by changes in professional application models, personal interests and desires and technological advances in mobile devices, multi-persona has emerged recently to keep balance between different aspects, in our daily life, on a single mobile terminal. In this context, mobile virtualization technology has turned the corner and currently heading towards widespread adoption to realize multi-persona. Although recent lightweight virtualization techniques were able to maintain balance between security and scalability of personas, the limited CPU power and insufficient memory and battery capacities, still threaten personas performance and viability. Throughout the last few years, cloud computing has cultivated and refined the concept of outsourcing computing resources, and nowadays, in the coming age of smartphones and tablets, the prerequisites are met for importing cloud computing to support resource constrained mobiles. From these premises, we propose in this paper a novel offloading-based approach that based on global resource usage monitoring, generic and adaptable problem formulation and heuristic decision making, is capable of augmenting personas performance and viability on mobile terminals. The experiments show its capability of reducing the resource usage overhead and energy consumption of the applications running in each persona, accelerating their execution and improving their scalability, allowing better adoption of multi-persona solution.

OPTiC: Optimizing Collaborative CPU–GPU Computing on Mobile Devices With Thermal Constraints

The CPU-graphic processing unit (GPU) co-execution of computation kernels on heterogeneous multiprocessor system-on-chip can significantly boost performance compared to the execution on either the CPU or the GPU alone. However, engaging multiple on-chip compute elements concurrently at the highest frequency may not provide the optimal performance in a mobile system with stringent thermal constraints. The system may repeatedly exceed the temperature threshold necessitating frequency throttling and hence performance degradation. We present OPTiC, an analytical framework that given a computation kernel can automatically select the partitioning point and the operating frequencies for optimal CPU–GPU co-execution under thermal constraints. OPTiC estimates, through modeling, CPU and GPU power, performance at different frequency points as well as the performance impact of thermal throttling and memory contention. Experimental evaluation on a commercial mobile platform shows that OPTiC achieves an average 13.68% performance improvement over existing schemes that enable co-execution without thermal considerations.

Striving for performance of discrete optimisation via memetic agent-based systems in a hybrid CPU/GPU environment

Agent-based memetic computational systems have been proven to be an efficient concept for solving complex computational problems. In this paper, we propose and evaluate a new hybrid architecture for such systems that effectively utilise CPU and GPGPU processing power by delegating parallel part of computations to GPGPU. The paper illustrates how to design and implement particular parts of the systems and discusses required conditions for evolutionary parameters and memetic operators. The concept is illustrated with one of the complex discrete problems – LABS.

Security of 6LoWPAN IoT Networks in Hospitals for Medical Data Exchange

It is well known that available approaches to IoT networks eschew certain security issues that may expose sensitive data to unauthorized entities. This paper proposes a security protocol for the IoT based communication. This protocol comprises of three levels of security checks namely authentication, confidentiality and integrity. Such algorithms do require large storage capacity and processing powers. However, nodes in IoT network have limited storage capacity and processing power. Proposed algorithm overcome this problem by reducing the number computations performed without losing the accuracy of data. Trust, Security and Privacy is established over a 6LoWPAN network running on RPL by keeping the CPU power consumption and storage requirements under check. The implementation of the proposed security scheme has been demonstrated for IoT based hospital networks in simulation with Cooja Network Simulator on Contiki OS

The INFN scientific computing infrastructure: present status and future evolution

The INFN scientific computing infrastructure is composed of more than 30 sites, ranging from CNAF (Tier-1 for LHC and main data center for nearly 30 other experiments) and nine LHC Tier-2s, to ∼ 20 smaller sites, including LHC Tier-3s and not-LHC experiment farms. A comprehensive review of the installed resources, together with plans for the near future, has been collected during the second half of 2017, and provides a general view of the infrastructure, its costs and its potential for expansions; it also shows the general trends in software and hardware solutions utilized in a complex reality as INFN. As of the end of 2017, the total installed CPU power exceeded 800 kHS06 (∼ 80,000 cores) while the total storage net capacity was over 57 PB on disk and 97 PB on tape: the vast majority of resources (95% of cores and 95% of storage) are concentrated in the 16 largest centers. Future evolutions are explored and are towards the consolidation into big centers; this has required a rethinking of the access policies and protocols in order to enable diverse scientific communities, beyond LHC, to fruitfully exploit the INFN resources. On top of that, such an infrastructure will be used beyond INFN experiments, and will be part of the Italian infrastructure, comprising other research institutes, universities and HPC centers.

INFN Tier–1: a distributed site

The INFN Tier-1 center at CNAF has been extended in 2016 and 2017 in order to include a small amount of resources (∼ 22 kHS06 corresponding to ∼ 10% of the CNAF pledges for LHC in 2017) physically located at the Bari-ReCas site (∼ 600 km distant from CNAF). In 2018, a significant fraction of the CPU power (∼ 170 kHS06, equivalent to ∼ 50% of the total CNAF pledges) is going to be provided via a collaboration with the PRACE Tier-0 CINECA center (a few km from CNAF), thus building a truly geographically distributed (WAN) center. The two sites are going to be interconnected via an high bandwidth link (400-1200 Gb/s), in order to ensure a transparent access to data residing on CNAF storage; the latency between the centers is small enough not to require particular caching strategies. In this contribution we describe the issues and the results of the production configuration, focusing both on the management aspects and on the performance provided to end-users.