The performance monitoring of computer systems is a complex affair, made even more challenging by the increasing gap between hardware and software. Methods that collect and feed data to performance analysis can usually be classified into one of two groups. Methods in the first group (e.g., software instrumentation), can be accurate but have unacceptable overheads and offer no insight into the interaction of the software and the hardware. Methods in the second group (e.g., hardware supported) run in real time and connect code to the response in the architecture, at a considerable decrease in accuracy, with a high degree of specificity to the hardware and therefore with limited actionable information made available. In this work, we focus on the latter group - the collection and analysis of raw performance data sourced from hardware Performance Monitoring Units, built into the majority of processors. The periodic collection of samples (Event Based Sampling), containing code locations, for instance, can be triggered on events such as retired instructions or cache misses. We quantify accuracy improvements to existing methods, propose a better performing method of performance data collection and a new architecture-agnostic method of performance data analysis. In our experiments, we analyze compiled, large-scale production workloads, industry standard benchmarks, and micro-benchmarks reproducing specific architectural behaviors. First, we focus on performance data collection methods and their accuracy, when establishing instruction retirement rates. These vary from least to most advanced and we examine the low level factors influencing accuracy. In particular, we employ a method based on Last Branch Records, a hardware facility present on some Intel architectures, which allows for the sampling of short records of last taken branches leading up to the sample. We compare the most commonly used method, its improved derivatives and the Last Branch Record based method. With respect to the first method, we observe accuracy improvements of up to 18x on synthetic benchmarks and 15x on real applications. Second, to further improve accuracy, we propose a new collection method named Hybrid Basic Block Profiling, fusing those based on Last Branch Records and the widely used Event Based Sampling. We apply simple machine learning techniques to determine the operating criteria. As a demonstration of capability, our method and a profiling tool are used to generate Instruction Mixes, which traditionally require good quality data at the most detailed level of granularity - basic blocks. Compared to software instrumentation, we observe an improvement in runtime of up to 76x, while keeping instruction attribution errors at 2.1% on average. In our third contribution we describe an architecture-agnostic performance analysis methodology called Hierarchical Cycle Accounting. The user is presented with a navigable hierarchy of microarchitectural issues, expressed in a metric that is universal and simple to compare - core cycles. We develop a reference implementation for the Intel Ivy Bridge microarchitecture and provide pointers for implementations on other processor architectures. Our approach and tool are successfully used by non-experts to improve and vectorize complex scientific code. Experts benefit as well - by being able to pinpoint software issues undetected by other methods, and identifying a bug in a family of Intel processors.
1Abass Alavi: A giant in Nuclear Medicine turns 80 and is still going strong!
Abass Alavi: A giant in Nuclear Medicine turns 80 and is still going strong!
A COMPARATIVE STUDY OF JOB MOTIVATION FACTORS BETWEEN EMPLOYEES AND OUTSOURCING STAFFIN PT. BANK NEGARA INDONESIA MAIN BRANCH OFFICE MANADO
Banking is one of the financial institutions, the actuator driving the growth of the economy that cannot be separated from development. There are two types of worker in banking company based on status, the first is employee and the second is outsourcing staff. Employees are people who are working for the company until their retired. And it has difference with outsourcing staff. An Outsourcing staff is not permanent employee and has limited term of work and working in a pre-determined time. Because of that, there are differences of job motivation between two status of workers.This research is a comparative study that uses a comparison method. The main object of this research is to compare the job motivation factors as achievement, recognition, work itself, responsibility, advancement, and personal growth between employees and outsourcing staff. The sample of this research is employees and outsourcing staff who working in Bank BNIMain Branch Office Manado as many as 121 respondents consist of 78 respondents for employees and 43 respondents for outsourcing staff of proportionate stratified random sampling.The result of this research showed there are no significant differences in job motivation factors of recognition, responsibility and personal growth. Furthermore, other results showed that there are significant differencesin job motivation factors of achievement, work itself, and advancement between employees and outsourcing staff in Bank BNI Main Branch Office Manado Keywords: job motivation, achievement, recognition,responsibility, personal growth
Read more1The Basic Block Reassembling Instruction Stream Buffer with LWBTB for X86 ISA
The potential performance of superscalar processors can be exploited only when processor is fed with sufficient instruction bandwidth. The front-end units, the Instruction Stream Buffer (ISB) and the fetcher, are the key elements for achieving this goal. Current ISBs could not support instruction streaming beyond a basic block. In x86 processors, the split-line instruction problem worsens this situation. In this paper, we proposed a basic blocks reassembling ISB. By cooperating with the proposed Line-Weighted Branch Target Buffer (LWBTB), the proposed ISB can predict upcoming branch information and reassemble current cache line together with the other cache line containing instructions for the next basic block. Therefore, the fetcher could fetch more instructions in a cycle with the assistance of the proposed ISB. Simulation results show that the cache line size over 64 bytes has a good chance to let two basic blocks present in a reassembled cache line and the fetch efficiency is about 90% as the fetch capacity is under 6.
Read moreJohn Foster McCreight : first premier of British Columbia
John Foster McCreight is not well known in the story of British Columbia, yet he held two of the highest offices – that of Premier and Judge of the Supreme Court. The reasons for this obscurity lie partly in the lack of personal records, and partly in the character of the man. He was of a retiring disposition, more concerned with the formation of statutes and correct legal interpretation than with more popular matters. He was not a man of his times because he stood for learning and principle in a period when action and expediency were more evident. His memory is over-shadowed by that of Amor De Cosmos and Judge Begbie, both of whom were more vivid personalities. McCreight was born in Northern Ireland in 1827, the son of a clergyman and related to John Foster, the last speaker of the Irish House of Commons, through his mother. He was educated at Trinity College, Dublin, and called to the Irish Bar in 1852. In 1853 he went to Australia, and practiced law in Melbourne where two of his cousins, Sir William Foster Stawell and Vesey Fitzgerald Foster held prominent positions. From Australia he came to Vancouver Island in 1860, and practiced law in Victoria. He lived in that city until 1880 where he spent nearly two years at Richfield in the Cariboo. In 1883 he moved to New Westminster where he lived until his retirement in 1897. He left Canada, proceeding first to Rome, then to Hastings, England, where he died in 1913. McCreight’s first interest in politics was shown in 1868 when he joined the Confederation League. It is believed that he was not actively interested in the struggle for responsible government in British Columbia, but that he was anxious to see a firm foundation of government laid when the new province entered Confederation. In 1871, Joseph Trutch, the Lieutenant-Governor chose McCreight as Premier, with a Cabinet consisting of Alexander Rocke Robertson, George Anthony Walkem, and Henry Holbrook. Thus McCreight became the first premier of the province and was responsible for passing a number of fundamental laws. Owing to his lack of interest in the idea of responsible government, he was defeated in the Legislature in 1872, and resigned as Premier. He remained a private member until 1875, but did not seek re-election. From 1875 to 1880 he devoted himself to an increasingly important law practice, and was appointed Queen’s Counsel. Some of his earlier law cases are of great general interest, notably that of “The Theatre Rumpus” in 1861, and Cranford versus Wright in 1862. The latter case brought McCreight into conflict with Judge Begbie over the conduct of legal affairs. In 1880, McCreight was made a puisne judge, and districted to Richfield. He found few cases to try there, and joined in the agitation against judges of the Supreme Court having to reside in their districts. From 1883 to 1897 he was resident judge at New Westminster, trying a great variety of cases, and proving himself to be the most learned of all judges in the province. McCreight was originally a strong adherent of the Anglican Church and…
Read moreHigh-bandwidth trace collection for multicomputer performance monitoring
Developing parallel applications requires sophisticated tools for tuning program performance. These tools require large quantities of data to be collected for analysis. A design for the collection and storage of trace data is presented in this paper. The design provides sufficient bandwidth and storage capacity for the collection of performance data from a 32-processor parallel computer using hybrid performance monitoring. A custom token-ring network is used to collect performance traces at a maximum bandwidth of 20 MBytes/sec. A 16 GByte disk array is utilized in the design for the storage of collected trace data. The resulting design provides a processor-independent method for collecting performance data from moderately parallel computing systems. >
Read more2ARMINTEL: A Heterogeneous Microprocessor Architecture Enabling Intel Applications on ARM
Intel is among the most popular microprocessor architectures introduced till to date. An extensively large number of applications have been developed on Intel ISA over the last more than three decades. However, Intel architectures also suffered intentionally or unintentionally from the cyber security related issues like Spectre, Meltdown and ZombieLoad. As compared to ARM & MIPS architectures, Intel is strictly proprietary; therefore it gives only minimal provision to implement safeguards against security threats at architectural level by its large scale users. In this work, we propose a heterogeneous microprocessor architecture named as ARMINTEL derived from our base concept of ARM inside Intel. Armintel uses an Intel front-end and an ARM back-end in its microarchitecture to make user able to run the applications developed for Intel in a secure execution environment of ARM. The contributions of this work are quad-fold starting from proposal on a heterogeneous microprocessor architecture, establishing its feasibility by driving Intel ISA mappings on ARM ISA, development of a complete simulation framework for functional evaluations of Armintel and last but not least is the performance estimation by adding cycle level measurements in the simulator. The results show that this new architecture can help replace security hazardous of Intel regime to a safer option of using open source ARM cores with front-end design to accept Intel binaries.
Read moreA Non-Linear GPU Performance Modeling Approach and Consolidated Linear Hardware Model Performance Evaluation of the LEAP Cluster
This document discusses the evaluation of a consolidated linear performance model of Texas State University's HPC LEAP Cluster, and the development of a non-linear model to represent the performance of the NVIDIA V100 Graphics Processing Unit (GPU). The consolidated model constructed for the LEAP Cluster was evaluated by using it to compute execution times for different sizes of processes and comparing those times to the actual processes run on LEAP. The model itself is found to effectively represent the performance trends of LEAP but fail to match the actual value. The GPU's performance analysis was done by first collecting performance data then using non-linear methods to model it. The number of floating-point operations required to perform a square matrix multiplication is readily known, so the performance data was gathered by running a simple square matrix multiplication program and measuring its execution time. The size of the matrices was then used to determine the number of floating-point operations and divided by the execution time to calculate FLOPS. The final model for the GPU was generated by using a set of logarithmic functions that mapped the GPU performance to a semi-logarithmic scale of the number of floating-point operations.
Read more358Collecting Performance Data with PAPI-C
Modern high performance computer systems continue to increase in size and complexity. Tools to measure application performance in these increasingly complex environments must also increase the richness of their measurements to provide insights into the increasingly intricate ways in which software and hardware interact. PAPI (the Performance API) has provided consistent platform and operating system independent access to CPU hardware performance counters for nearly a decade. Recent trends toward massively parallel multi-core systems with often heterogeneous architectures present new challenges for the measurement of hardware performance information, which is now available not only on the CPU core itself, but scattered across the chip and system. We discuss the evolution of PAPI into Component PAPI, or PAPI-C, in which multiple sources of performance data can be measured simultaneously via a common software interface. Several examples of components and component data measurements are discussed. We explore the challenges to hardware performance measurement in existing multi-core architectures. We conclude with an exploration of future directions for the PAPI interface.
Read more2An Approach for Attainment of Relevant Program Outcomes through Microcontroller Laboratory Course
Embedded product design has emerged as an important area of study in the curriculum for undergraduate program in Computer Science. This necessitates the inclusion of hardware courses like Digital Electronics, Computer Organization and Microcontrollers as pre-requisite courses, in the lower semesters. This helps students to develop keen interest, skill and proficiency in the area of Embedded system design. This also helps them to compete with peers from other engineering branches in the domain of Embedded Systems. Here we present a set of activities, their related outcome based assessment techniques and outcome based strategies applied to the laboratory course on 8051 Microcontroller laboratory in Computer Science&Engineering at the IV Semester level. This laboratory was an extension of the hands on experience the students had in the course on Digital Electronics in the III Semester. The laboratory course on 8051 Microcontrollers was based on the usage of 8051 compatible Atmel 89C51ED2. The course was designed to consist of initially, conducting simple programming exercises using 8051 assembly language and then simple interfacing exercises using Embedded C programming. Later when they developed some proficiency and familiarity in 8051 environment student' teams were made to work on course projects which required them to design and build projects using 8051 or any other microcontroller of their choice. This promoted their self-learning, improved their knowledge of programming the microcontroller to interface with the outside world to develop useful applications and provide solutions to some common problems in society. This activity resulted in enhanced motivational levels amongst students, increased their involvement in the team and improved their knowledge due to self learning.
Read more水稻穗著粒密度之研究I.穗著粒密度類型之分類、預測模式之建立及其相關性狀之探討
Using 34 rice cultivars/lines as materials, the principal component analysis was performed to elucidate the relationships between the panicle density parameters and its 10 related characters. Results showed that the first principal component (Z1) was mainly concerned with the length factor, the primary branch factor and the total number of spikelets per pancile; The second principal component (Z2) was largely due to the pancile weight, the primary branch factor and the total number of spikelets per panicle; The third principal component (Z3) was mainly associated with the length of rachis, the number of primary branches, the number of secondary branches and the ratio of the total number of spikelets of secondary branches and the total number of spikelets per panicle; while, the fourth principal component (Z4) was mainly performed by the pancile length, the number of primary branches and the average number of spikelets per primary branch. The relationships between the four principal components and panicle density parameters, evaluated by two different methods K (Number of spikelets/Total length of rachis and primary branches in cm) and L (Number of spikelets/10cm panicle length), were established by stepwise regression analysis method, A total eight equations were obtained, respectively. Among them, four for panicle density parameter K and two for panicle density parameter L, their coefficients of multiple determination were very high (＞0.84). Equations, K2 (=0.918Z1-0.215Z3, R^2=0.888) and L3 (0.808Z1+0.374Z2-0358Z4, R^2=0.920), estimated by the use of the least principal components, respectively, were selected as the best models for predicting the panicle density rice. Moreover, from the relationships between panicle density parameters and its related characters, it was found that the higher panicle weight, average number of spikelets of primary branches, secondary branch factor and the total number of spikelets per panicle. the higher their pancile density. While, the panicle length, length of rachis and the sum of the length of primary branches were down lower as their pancile density toward high. From the scatter diagram for the related characters of panicle density, 34 cultivars/lines, belonging to seven groups, could be classified into four different panicle density types: very low, low, middle and high panicle density type.
Read moreاثر روش های مختلف خاکورزی و کنترل علف های هرز بر عملکرد پنبه(Gossypium hirsutum)
به منظور ارزیابی تاثیر استفاده توام از روش مکانیکی (روش های متفاوت خاکورزی) و کنترل شیمیایی علف های هرز بر عملکرد پنبه، آزمایشی در سال 1391 در شهرستان بشرویه با آرایش اسپیلت پلات بر پایه طرح بلوک های کامل تصادفی به اجرا درآمد. فاکتور های آزمایش شامل روش خاکورزی پنبه به عنوان فاکتوراصلی در 3 سطح، خاکورزی مرسوم، خاکورزی مستقیم بدون شخم و خاکورزی حفاظتی و روش های مختلف کنترل شیمیایی علف های هرز در 5 سطح (علف کش تری فلوکسی سولفورون سدیم به میزان 10، 15 و 20 گرم در هکتار، به همراه تیمارهای وجین و عدم کنترل علف های هرز) به عنوان فاکتور فرعی در 4 تکرار انجام شد. نتایج نشان داد که بین روشهای خاکورزی اختلاف معنی داری وجود داشت به طوری که عملکرد وش، طول شاخه رویا، تعداد شاخه رویا و زایا، ارتفاع بوته و تعداد قوزه در سیستم خاکورزی حفاظتی و سیستم بدون شخم بیشتر از تیمارخاکورزی مرسوم بود. در بین دُز های علفکش انوک، استفاده از دُز 20 گرم در هکتار منجر به ایجاد بیشترین مقدار در صفات اندازه گیری شده گردید. اثر متقابل خاکورزی و کنترل علف های هرز بر تعداد شاخه رویا و طول شاخه رویا معنی دار بود. بیشترین تعداد شاخه رویا به ترتیب در خاکورزی حفاظتی و وجین علف های هرز و به تعداد 59/2 و 31/2 بدست آمد. بیشترین طول شاخه رویا در خاکورزی حفاظتی (79/15 سانتی متر) و در تیمار وجین در کنترل علف هرز (23/15سانتی متر) مشاهده شد.
Read moreThe anatomical basis and clinical application of free wrist crease flap
Objective To summarize the cutaneous branches of the superficial branch of radial artery distribution were dissected and to discuss the operational technique and clinical efficacy of repairing hand tissue defect by the flap based on the superficial branch of radial artery. Methods From May, 2014 to February, 2017, 8 fresh hand specimens perfused with red latex to observe the distribution and characteristics. And 24 cases of fingers tissue defects (3 thumbs, 6 index fingers, 5 middle fingers, 9 ring fingers, and 1 litter finger) which were repaired with the free wrist crease flap based on the cutaneous branches of superficial palmar branch of the radial artery. Combined fracture in 5 cases. Combined rupture or defect of the flexor tendon in 8 cases. Combined vascular nerve injury in 6 cases. The maximum area of the flap was 5.5 cm × 3.2 cm, and the minimum was 2.0 cm × 1.5 cm. Regular followed-up was performed after operation. Results Superficial palmar branch of the radial artery came from the level of radial styloid process ulnar palmar then through ulnar of tubercle of scaphoideum and tended to the thenar muscle. It gave 1-3 branches of the direct skin perforators at distal wrist crease and the anatomical location was constant. The vascular pedicle length could achieve 2.0-3.2 cm. All flaps were survived uneventfully. Twenty-four cases were followed-up for 6-29 months (mean, 18.6 months), and the appearance and functional recovered satisfactorily. Flap sensation recovered to S3+. The 2 point discrimination was 8-13 mm. There was no complication in donor area. According to the fingers total active motion (TAM) method to assess: excellent in 17 cases, good in 5 cases, fair in 2 cases. The excellent and good rate was 91.7%. Conclusion The wrist crease cutaneous branches of superficial palmar branch of the radial artery anatomical location is constant. And the flap based on the artery has a reliable blood supply. The vascular pedicle flap is long enough and the operational technique is easy and safe. The postoperative effect is satisfied. It is an ideal method for repairing finger tissue defect. Key words: Superficial palmar branch of the radial artery; Wrist crease flap; Fingers injury; Tissue defect; Repair
Read more3Multi-Objective Optimisation Method for Identifying Retired Points of Electric Vehicle Batteries
A multi-objective optimisation method to quantitatively identify retired points of electric vehicle (EV) batteries is proposed to minimise the life cycle cost (LCC) of EV batteries and the total annual cost (TAC) of energy storage systems (ESS). It features comprehensive considerations of battery capacity degradation characteristics and energy storage capacity optimisation. The effectiveness of the proposed method is demonstrated by a case study. The influence of the purchase cost and the profit of batteries in the second life are analysed. The Pareto front of LCC and TAC is given. The trade-off point is obtained when SOHre is 0.855 and the corresponding LCC and TAC are 28,742.2 USD and 7,905.5 USD. Results indicate that retired points are intensively dependent upon the optimal capacity, LCC and TAC. Both LCC and TAC can be reduced by decreasing the purchase cost and increasing the profit without changing the retired points.
Read more5Applied anatomy of the dorsal branch of the proper digital nerve
Objective To provide anatomic basis for the clinic application of the dorsal branch of the proper digital nerve. Methods 48 fingers of 12 hands of adult cadavers were dissected. The dorsal branch of the proper digital nerve were observed and measured about the rate of appearance, diameter, course and branches.Results The dorsal branch of the proper digital nerve was seen originating from the proper digital nerve on both sides of the index, middle and ring fingers and from the nerve on radial side of the little finger in 92.86 % of the specimens (78/84 sides). The absent rate of the dorsal side of the little finger in 92.86 % of the specimens (78/84 sides). The absent rate of the dorsal branch was 7.14 % (6/84 sides). The dorsal branch began at the level of proximal - mid third of the proximal phalanx of the finger. It coursed obliquely and distally from lateral aspect of the finger to the dorsal aspect. At dorsal - radial margin or dorsal - ulnar margin of PIP joint, 2 to 5 branches were sent off to the dorsum of the middle phalanx. Except the little finger, the diameter of the radial dorsal branch and the ulnar dorsal branch had significant difference, the former being thicker than the latter. The ulnar dorsal branch of the little finger was absent. Conclusions The sourse of the dorsal branch of the proper digital nerve of index, middle and ring fingers on both sides, and that of the little finger on the radial side is constant. It can be harvested with the reverse digital island flap to repair finger pulp defect.
Read moreClinical significance of reserving the vagus nerve celiac branch and hepatobiliary branch in radical gastrectomy
Objective To investigate the clinical significance of reserving the vagus nerve celiac branch and hepatobiliary branch in radical gastrectomy. Methods Fifty-seven cases of gastric cancer from January 2006 to January 2012 were randomly divided into two groups,reserving the vagus nerve ce-liac branch and hepatobiliary branch group（group A）and not reserving vagus nerve group（group B）. The relevant data of the two groups were compared,such as operation time,volume of operation bleed-ing,the number of lymph node dissection,the rate of complication,the time of first exhaust,defecation and obvious discomfort after meals,clinical symptoms,the results of electronic gastroscopy,incidences of gallstones and recurrence metastasis in postoperative follow-up of 3,6,12 and 24 months and 2 - year life quality. Results The operation time in group A was significantly longer than that in group B,and the volume of bleeding in group A was more than that in group B（P ﹤ 0. 05）. The number of lymphade-nectomy in the two groups was 18. 02 and 18. 38（P ﹥ 0. 05）;the rate of postoperative complications was 16. 67%（6 / 36）and 19. 05%（4 / 21）,P ﹥ 0. 05;the times of first exhaust,defecation in group A were earlier than those in group B（P ﹤ 0. 01）;the life quality in group A was significantly better than that in group B（P ﹤ 0. 01）;the 2 - year survival rates in two groups were both 100% . Conclusions Reserving the vagus nerve celiac branch and hepatobiliary branch in radical gastrectomy can speed up the recovery of gastrointestinal function and better absorption of nutrients,in the case of not affecting the rad-ical effect,the life quality of patients is improved. Key words: Radical gastrectomy; Reserving the vagus nerve; Hepatobiliary branch; Celiac branch; Life quality
Read more
Copyright 2026 Cactus Communications. All rights reserved.
