Search or add a thesis

Advanced Search (Beta)
Home > Power Efficient Resource Allocation in High Performance Computing Systems

Power Efficient Resource Allocation in High Performance Computing Systems

Thesis Info

Access Option

External Link

Author

Hussain, Hameed

Program

PhD

Institute

COMSATS University Islamabad

City

Islamabad

Province

Islamabad.

Country

Pakistan

Thesis Completing Year

2016

Thesis Completion Status

Completed

Subject

Computer Science

Language

English

Link

http://prr.hec.gov.pk/jspui/bitstream/123456789/7997/1/Hameed%20Hussain.pdf

Added

2021-02-17 19:49:13

Modified

2024-03-24 20:25:49

ARI ID

1676727812166

Asian Research Index Whatsapp Chanel
Asian Research Index Whatsapp Chanel

Join our Whatsapp Channel to get regular updates.

Similar


Power Efficient Resource Allocation in High Performance Computing Systems An efficient resource allocation is a fundamental requirement in High Performance Computing (HPC) systems. Many projects are dedicated to large-scale distributed computing systems that have designed and developed resource allocation mechanisms with a variety of architectures and services. Resource allocation mechanisms and strategies play a vital role towards the performance improvement of all the high performance computing classifications. Therefore, a comprehensive discussion of widely used resource allocation strategies deployed in distributed high performance computing environment is required. The author has classified the distributed high performance computing systems into three broad categories, namely: (a) cluster, (b) grid, and (c) cloud systems and defines the characteristics of each class by extracting sets of common attributes. All of the aforementioned systems are cataloged into pure software and hybrid/hardware solutions. The system classification is used to identify approaches followed by the implementation of existing resource allocation strategies that are widely presented in the literature. More computational power is offered by high performance computing systems to cope with CPU intensive applications. However, this facility comes at the price of more energy consumption and eventually higher heat dissipation. As a remedy, these issues are being encountered by adjusting system speed on the fly so that application deadlines are respected and also, the overall system energy consumption is reduced. In addition, the current state of the art of high performance computing, particularly the multi-core technology opens further research opportunities for energy reduction through power efficient scheduling. However, the multi-core front is relatively unexplored from the perspective of task scheduling. To the best of our knowledge, very little is known as of yet to integrate power efficiency component into real-time scheduling theory that is tailored for high performance computing particularly multi-core platforms. In these efforts the author first proposes a technique to find the least feasible speed to schedule individual tasks. The proposed technique is experimentally evaluated and the results show the supremacy of our approach over the existing counterpart called first feasible speed. However, this solution is at the cost of delayed response time. The experimental xi results are in accordance with the mathematical formulation established in this work. To minimize power consumption, the author did another attempt by applying genetic algorithm on first feasible speed. The newly proposed approach is termed as genetic algorithm with first feasible speed. The author compares the results obtained through aforementioned approach with existing techniques. It is worth mentioning that proposed technique outperforms first feasible speed and least feasible speed with respect to energy consumption and response time perspectives respectively. Load balancing is also vital for efficient and equal utilization of computing units (systems or cores). To load balancing among computing units, the author applies lightest task-migration (task-shifting) and task splitting mechanisms in a multi-core environment. In task shifting a task having minimum load on a highly utilized computing unit is fully transferred to a low utilized computing unit. In task splitting, the load of a task from a highly utilized computing unit is shared among the computing unit and a low utilized computing unit. It concludes from the given results that task splitting mechanism fully balance load but it is more time consuming as compared to task shifting strategy. Keywords: High performance computing, Real time system, Scheduling, Resource allocation, Resource management, Genetic algorithm, Task migration and Task splitting.
Loading...
Loading...

Similar Books

Loading...

Similar Chapters

Loading...

Similar News

Loading...

Similar Articles

Loading...

Similar Article Headings

Loading...

مسجد سلطان حسن اور رفاعی مسجد

مسجد سلطان حسن اور رفاعی مسجد

قلعے کے دائیں طرف سلطان حسن اور بائیں جانب رفاعی مسجد کی بلند و بالا عمارات دامنِ دل کھینچتی ہیں ۔ان دونوں مساجد کے درمیان ایک کشادہ شاہرہ ہے جوان مساجد کے آخر میں ایک وسیع و عریض باغیچے میں اختتام پذیر ہوتی ہے ۔راستے کے دو نوں جانب دو فٹ چوڑی اور اتنی ہی اونچی دیوار ہے جس نے راستے کی حنا بندی کی ہوئی ہے ۔دکتورہ بسنت نے اپنے پرس سے ایک چادر نکالی اور اس دیوار پر بچھا دی ۔اس کے بعد انھوں نے ڈبل روٹی ، کھیرے ،پنیر اور جیم کی ایک ڈلی چادر کے اوپر چن دی ۔پرس کے اندرونی جیب سے چھری نکالی ڈبل روٹی کو درمیان سے کاٹا اس کے اندر کھیرے اور پنیر کو جیم لگا کر رکھ دیا اور ہمارے سامنے پیش کر دیا ۔دکتورہ بسنت کی اس پرخلوص دعوتِ شیراز پر کون نہ مرتا۔

مصری عورت فرعون کے زمانے سے تاحال بااختیار ہے ۔امورِخانہ داری سے امورِ حکمرانی تک یہ باہمت عورت پدر سری معاشرے میں بھی اپنے حقوق اور فرائض سے لطف اندوز ہوتی رہی ہے ۔دعوتِ شیراز کے بعد ہم الرفاعی مسجد کی طرف روانہ ہوئے تو ایک خاتون اور دو مرد ہماری طرف آئے اور مجھ سے پوچھا کہ آپ ہندوستان سے آئے ہیں ۔میں چونکہ شلوار قمیض میں تھا اس لیے ان کو میرے مصری نہ ہونے پر یقین ہو گیا۔میں نے کہا میں پاکستان سے ہوں ۔انھوں نے مجھے چالیس جنین کی ٹکٹ پکڑا دی تو میرے میزبانوںکو مصری محکمہ سیاحت والوںکی یہ حرکت بری لگی ۔احمد نے ان سے یہ کہا کہ یہ مسلمان ہیں ایک مسلمان کے مسجد میں داخل ہو نے پر آپ ان سے ٹکٹ لیں گے ؟انھوں نے کہا یہاں لوگ نماز توپڑھتے نہیں تاریخ...

مختلف ادیان میں سزاؤں کا تصور اور اسلامی حدود پر غیر انسانی ہونے کے الزام کا تحقیقی جائزہ

Many people propagate about the Islamic Hudūd i.e. Punish against crimes stated in the Holy Quran. They are of the view that the Islamic Hudūd punishments are more severe, cruel a brutal. Although it a propaganda against Islam. This negative propaganda is an obstacle, hindrance in the way of enforcement of Hudūd. If the Islamic Hudūd are compared with the other Religions, punishments for different crimes, which have been mentioned in their books and Religious history, they are either similar or more sever and brutal than Islamic Hudūd, i.e. Punishment. It show that more sever & cruel punishments against crimes were present in heavenly and man-made Religious before Islam. The research under consideration is about the comparison and contrast between Islamic punishments and the punishments present the international or worldly Religions so that it may be cleared to the whole world that only Islam is a Religion in the world that no other Religion can compete regarding prevention of crimes. It will highlight the sublimity and loftiness of Islam and also make it clear the fact of hollow, attractive slogans of the present modern time.

Studies on Blood Flow Indices Using Doppler Ultrasonography During Estrous Cycle in Sahiwal Cows

The Sahiwal cow is among one of the well-known dairy breed of zebu (humped) cattle, representing Bos indicus. Information on reproductive aspects is lacking in this breed. Therefore, the intentions of the first experiment was to determine the variations in flow of blood of the uterine artery (UA) based on RI and PI using color-Doppler ultrasonography, during the entire estrous cycle in Sahiwal cattle. Additionally, we characterized the ovarian follicular and luteal dynamics, and the correlation of uterine blood flow, with P4 concentrations in systemic blood. Nine (n = 9), adult, healthy, cycling, lactating Sahiwal cows were registered in the study at their spontaneous estrous. On subsequent estrus, all cows were subjected to AI twelve hours after the onset of standing heat with semen that was freeze, and pregnancy was detected on day thirty post AI. Each cow examined transrectally with B-mode and color-Doppler ultrasound on alternate days throughout the estrous cycle by the same operator. For B-mode, comprehensive pictures of the ovaries were drawn to measure the quantity, diameter, and comparative position of the follicle and CL. The day on which signs of estrus and dominant ovulatory follicle observed was carefully chosen as -1 Day, whereas the day on which dominant follicle (DF) vanish referred as Day 0. Doppler measurements were determined from flow of blood in both the left and right uterine arteries. The analysis was based on Doppler spectrum. The RI and PI were measured to reflect changes in blood flow. RI and PI were calculated from the built-in caliper for measures of blood flow parameters, which were displayed on monitor. Values were recorded when minimum three analogous and uninterrupted waveforms were noted, and three recordings of each day of each side were averaged. Mean RI readings were significantly higher (P < 0.05) at Day -1 (estrus) compared to Day 0 (ovulation). Similarly, mean RI values remained lower (P < 0.05) at Day -1 and Day 0 as compared to remaining days of estrous cycle. We observed a robust outcome of cattle and the days of entire heat cycle (P < 0.05) on RI readings but their interaction was not significant. Mean PI ix reading between -1 Day (estrus) and Day 0 (ovulation) did not differ (P > 0.05) and remained lower. Mean PI value was higher on Day 10 than on Day -1 and Day 0 (P < 0.05). We observed a robust effect of cattle and days of the entire heat cycle (P < 0.05) on PI readings but their interaction was not significant. Concentrations of P4 were greater on Day 12 (6.4 ± 1.0 ng/mL) and then started decreasing on Day 14 (5.0 ± 1.01 ng/mL) until reaching nadir. It is concluded that RI of uterine arteries, as a measure of blood flow, is considerably lower at estrus and ovulation, than during diestrus, whereas PI is substantially elevated during diestrus compared to heat and ovulation in Sahiwal cattle. The objectives of the 2nd study were to determine if there are changes in LS, progesterone (P4), and LBF between pregnant and non-pregnant Bos indicus dairy cattles during the first three weeks after insemination and if these parameters are related to each other. For this, Sahiwal cows (n = 13) of mixed parity (1 - 3) that were healthy, regularly cycling, and lactating, and of 430 ± 18 kg (mean ± SD) body weight were enrolled for the study. All cows were inseminated using frozen thawed semen 12 h after the start of standing estrus. B-mode ultrasonography was performed repeatedly with a 12 h interval to confirm the absence of POF (ovulation), and this day was referred to as Day 0 of the estrous cycle. All inseminated cows were retrospectively categorized as pregnant or nonpregnant. In order to compare the LS and LBF after AI, brightness mode and color-Doppler ultrasonography of the ovaries were performed on Days 4, 5, 6, 7 (first week); 8, 10, 12, 14, (second week); and 16, 17, 18, 19, 20, 21 (third week) in pregnant and nonpregnant cows. During the first and third weeks, ultrasonography was performed daily, while during the second week, cows were examined on alternate days. Thus, each cow had the ovaries ultrasound-mapped fourteen times, resulting in a total of 182 recordings from all cows. The stored images were then subjected to offline analysis using computer assisted image analysis software, x Image J (National Health Institute, Bethesda, MD, USA). In order to minimize the chances of error, mean values of the three images were recorded during each examination for both LS and LBF. Results revealed that the mean LS increased (P < 0.05) from Day 4 (0.9 cm2) to Day 7 (1.3 cm2) indicating the growth of luteal tissue during the first week. However, the average LS did not vary between Day 4 (first week) and Day 19 (third week). The P4 rose (P < 0.05) from Day 4 (1.7 ng/mL to Day 7 (3.0 ng/mL) during the first week, and increased (P < 0.05) further from Day 8 (3.9 ng/mL) to Day 14 (5.8 ng/mL). The LBF increased (P < 0.05) on Day 4 (0.3 cm2) to Day 7 (0.7 cm2), indicating a two-fold rise within 3 days of the first week. During the third week, LBF declined (P < 0.05) from 0.8 cm2 (Day17) to 0.2 cm2 (Day 21) in nonpregnant cows. The mean LS increased (P < 0.05) on Day 4 (1.0 cm2) to Day 7 (1.8 cm2) during the first week. In the following days, LS amplified (P < 0.05) from 1.9 cm2 (Day 8) to 2.8 cm2 (Day 21), indicating a consistent increase in luteal tissue during the first three weeks of pregnancy. The P4 showed a rise (P < 0.05) on Day 4 (1.4 ng/mL) Day 7 (2.6 ng/mL) and from 2.9 ng/mL (Day 8) to 8.4 ng/mL (Day 21). The LBF increased (P < 0.05) more than double from Day 4 (0.4 cm2) to Day 7 (0.9 cm2), as well as from Day 8 (1.0 cm2) to Day 21 (2.4 cm2) in pregnant cows. In conclusion, the present study has provided new information about relationships between LS, P4, and LBF, and indicated that LBF is a more sensitive parameter than LS and P4 to detect the differences of luteal function during the 1st three (3) weeks post AI in pregnant and nonpregnant Bos indicus dairy cattles. Furthermore, this approach could be effectively used to decrease the re-insemination interval, number of days open, and calving interval for the optimization of reproductive management in dairy cows. The comprehensive objectives of these studies are to make advancements in estrus and ovulation synchronization, re-synchronization, decreasing insemination and calving interval, early detection of pregnancy and fetal loss, and minimizing the pathological conditions of the uterus and ovary by using this novel and non-invasive technique.