An Experimental Time-Sharing System

Corbató, F. J.; Merwin-Daggett, Marjorie; Daley, Robert C.

doi:10.1007/978-1-4757-3510-9_8

Cited by 34 publications

(40 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…To this end, we adopt the Multi-level Feedback Queue (MLFQ) [10,27] to implement our ACK scheduling at the receiver side. To enforce MLFQ, PAC introduces N distinct queues to store ACK packets, which are assigned with N different priority levels.…”

Section: How To Schedule Ack Packets?mentioning

confidence: 99%

PAC: Taming TCP Incast Congestion Using Proactive ACK Control

Chen

Lan

et al. 2014

2014 IEEE 22nd International Conference on Network Protocols

View full text Add to dashboard Cite

Abstract-TCP incast congestion which can introduce hundreds of milliseconds delay and up to 90% throughput degradation, severely affecting application performance, has been a practical issue in high-bandwidth low-latency datacenter networks. Despite continuous efforts, prior solutions have significant drawbacks. They either only support quite a limited number of senders (e.g., 40-60), which is not sufficient, or require non-trivial system modifications, which is impractical and not incrementally deployable.We present PAC, a simple yet very effective design to tame TCP incast congestion via Proactive ACK Control at the receiver. The key design principle behind PAC is that we treat ACK not only as the acknowledgement of received packets but also as the trigger for new packets. Leveraging datacenter network characteristics, PAC enforces a novel ACK control to release ACKs in such a way that the ACK-triggered in-flight data can fully utilize the bottleneck link without causing incast collapse even when faced with over a thousand senders.We implement PAC on both Windows and Linux platforms, and extensively evaluate PAC using small-scale testbed experiments and large-scale ns-2 simulations. Our results show that PAC significantly outperforms the previous representative designs such as ICTCP and DCTCP by supporting 40X (i.e., 40→1600) more senders; further, it does not introduce spurious timeout and retransmission even when the measured 99 th percentile RTT is only 3.6ms. Our implementation experiences show that PAC is readily deployable in production datacenters, while requiring minimal system modification compared to prior designs.

show abstract

Section: How To Schedule Ack Packets?mentioning

confidence: 99%

PAC: Taming TCP Incast Congestion Using Proactive ACK Control

Chen

Lan

et al. 2014

2014 IEEE 22nd International Conference on Network Protocols

View full text Add to dashboard Cite

show abstract

“…In decay-usage [Corbató et al, 1962], like user-fb, the user who has consumed the fewest resources is favored, but unlike strict user-fb, a user's resource usage is decayed over time. As resource usages decay, decay-usage behaves more like ps.…”

Section: Cluster Schedulingmentioning

confidence: 99%

“…Two-tiered batchactive scheduling is a restricted case of priority-based queues or classed schedulers [Corbató et al, 1962]. Classed schedulers pick from the highest non-empty class before proceeding to lower classes.…”

Section: Applying Priority-class Schedulingmentioning

confidence: 99%

Cluster scheduling for explicitly-speculative tasks

Petrou

Ganger

Gibson

2004

Proceedings of the 18th Annual International Conference on Supercomputing

View full text Add to dashboard Cite

Public reporting burden for the collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for Information Operations and Reports, 1215 Jefferson Davis Highway, Suite 1204, Arlington VA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to a penalty for failing to comply with a collection of information if it does not display a currently valid OMB control number. AbstractA process scheduler on a shared cluster, grid, or supercomputer that is informed which submitted tasks are possibly unneeded speculative tasks can use this knowledge to better support increasingly prevalent user work habits, lowering user-visible response time, lowering user costs, and increasing resource provider revenue.Large-scale computing often consists of many speculative tasks (tasks that may be canceled) to test hypotheses, search for insights, and review potentially finished products. For example, speculative tasks are issued by bioinformaticists comparing dna sequences, computer graphics artists rendering scenes, and computer researchers studying caching. This behaviorexploratory searches and parameter studies, made more common by the costeffectiveness of cluster computing -on existing schedulers without speculative task support results in a mismatch of goals and suboptimal scheduling. Users wish to reduce their time waiting for needed task output and the amount they will be charged for unneeded speculation, making it unclear to the user how many speculative tasks they should submit. This thesis introduces 'batchactive' scheduling (combining batch and interactive characteristics) to exploit the inherent speculation in common application scenarios. With a batchactive scheduler, users submit explicitlylabeled batches of speculative tasks exploring ambitious lines of inquiry, and users interactively request task outputs when these outputs are found to be needed. After receiving and considering an output for some time, a user decides whether to request more outputs, cancel tasks, or disclose new speculative tasks. Users are encouraged to disclose more computation because batchactive scheduling intelligently prioritizes among speculative and non-speculative tasks, providing good wait-time-based metrics, and because batchactive scheduling employs an incentive pricing mechanism which charges for only requested task outputs (i.e., unneeded speculative tasks are not charged), providing better cost-based metrics for users. These aspects can lead to higher billed server utilization, encouraging batchactive adoption by resource providers organized as either cost-or profit-centers. vi · Cluster sche...

show abstract

“…[2] As computers became more prevalent, scientists and technologists explored ways to make large scale computing power available to more users through time sharing, experimenting with algorithms to provide the optimal use of the infrastructure, platform and applications with prioritized access to the CPU and efficiency for the end users. [3] …”

Section: Introductionmentioning

confidence: 99%

Learning in the cloud: a new challenge for a global teaching system in optics and photonics

Sultana

Christ

Feisst³

et al. 2014

SPIE Proceedings

View full text Add to dashboard Cite

Nowadays, it is assumed of many applications, companies and parts of the society to be always available online. However, according to [Times, Oct, 31 2011], 73% of the world population do not use the internet and thus aren't "online" at all. The most common reasons for not being "online" are expensive personal computer equipment and high costs for data connections, especially in developing countries that comprise most of the world's population (e.g. parts of Africa, Asia, Central and South America). However it seems that these countries are leap-frogging the "PC and landline" age and moving directly to the "mobile" age. Decreasing prices for smart phones with internet connectivity and PC-like operating systems make it more affordable for these parts of the world population to join the "always-online" community. Storing learning content in a way accessible to everyone, including mobile and smart phones, seems therefore to be beneficial. This way, learning content can be accessed by personal computers as well as by mobile and smart phones and thus be accessible for a big range of devices and users. A new trend in the Internet technologies is to go to "the cloud". This paper discusses the changes, challenges and risks of storing learning content in the "cloud". The experiences were gathered during the evaluation of the necessary changes in order to make our solutions and systems "cloud-ready".

show abstract

An Experimental Time-Sharing System

Cited by 34 publications

References 4 publications

PAC: Taming TCP Incast Congestion Using Proactive ACK Control

PAC: Taming TCP Incast Congestion Using Proactive ACK Control

Cluster scheduling for explicitly-speculative tasks

Learning in the cloud: a new challenge for a global teaching system in optics and photonics

Contact Info

Product

Resources

About