Swami Sivasubramanian

Swami Sivasubramanian Swami Sivasubramanian is an influencer

Greater Seattle Area
112K followers 500+ connections

About

I run AI and Data services in AWS.

I have been awarded (or filed for) more than…

Articles by Swami

Activity

Join now to see all activity

Experience

Education

  • Vrije Universiteit Amsterdam (VU Amsterdam) Graphic

    Vrije Universiteit Amsterdam (VU Amsterdam)

    -

    - Published around 30 research papers in ACM/IEEE journals and conferences.
    - Won a best research paper award from IEEE Service Computing Technical committee.

  • -

    Activities and Societies: Teaching Assistant for Real-time systems, Research Assistant - worked on design and implementation of new RTOS scheduling algorithms, SITAR - Society for Indian Tradition and ARts

    - Passed with dual honors (research excellence and teaching excellence awards). GPA: 4.0/4.0

    - Involved in research of real-time scheduling for dynamic real-time systems. Areas focussed: Value-based Scheduling and Feedback-controlled real-time scheduling.
    - Published 2 journals and seven conference/workshop papers.
    - Awarded Research Excellence Award.

    - Taught "Digital Systems Design" and "Real-Time Systems" for undergraduate courses.
    - Awarded Teaching Excellence…

    - Passed with dual honors (research excellence and teaching excellence awards). GPA: 4.0/4.0

    - Involved in research of real-time scheduling for dynamic real-time systems. Areas focussed: Value-based Scheduling and Feedback-controlled real-time scheduling.
    - Published 2 journals and seven conference/workshop papers.
    - Awarded Research Excellence Award.

    - Taught "Digital Systems Design" and "Real-Time Systems" for undergraduate courses.
    - Awarded Teaching Excellence Award by President of ISU.
    - Was involved in building the practical lab exercises for Real-Time Systems Course (CprE 558 ) from scratch

  • -

Publications

  • Feedback control for real-time scheduling

    American Control Conference

    Most real-time scheduling algorithms are open-loop algorithms as the scheduling decisions are based on the worst-case estimates of task parameters. In recent years, the "closed-loop" scheduling has gained importance due to its applicability to many real-world problems wherein the feedback information can be exploited efficiently to adjust task and/or scheduler parameters, thereby improving the system's performance. In this paper, we discuss an open-loop dynamic scheduling algorithm that employs…

    Most real-time scheduling algorithms are open-loop algorithms as the scheduling decisions are based on the worst-case estimates of task parameters. In recent years, the "closed-loop" scheduling has gained importance due to its applicability to many real-world problems wherein the feedback information can be exploited efficiently to adjust task and/or scheduler parameters, thereby improving the system's performance. In this paper, we discuss an open-loop dynamic scheduling algorithm that employs a notion of task overlap in the scheduler in order to provide some flexibility in task execution time. Then we present a novel closed-loop approach for dynamically estimating the execution time of tasks based on both deadline miss ratio and task rejection ratio in the system. This approach is highly preferable for firm/soft real-time systems since it provides a firm performance guarantee in terms of deadline misses while achieving a high guarantee ratio. We design the proportional-integral controller and H∞ controller for closed loop scheduling. We evaluate the performance of the open-loop and the closed-loop approaches using simulation studies. We show that the closed-loop dynamic scheduling offers a better performance over the open-loop scheduling under all practical conditions.

    Other authors
    See publication
  • Amazon Dynamo

    -

    Reliability at massive scale is one of the biggest challenges we face at Amazon.com, one of the largest e-commerce operations in the world; even the slightest outage has significant financial consequences and impacts customer trust. The Amazon.com platform, which provides services for many web sites worldwide, is implemented on top of an infrastructure of tens of thousands of servers and network components located in many datacenters around the world. At this scale, small and large components…

    Reliability at massive scale is one of the biggest challenges we face at Amazon.com, one of the largest e-commerce operations in the world; even the slightest outage has significant financial consequences and impacts customer trust. The Amazon.com platform, which provides services for many web sites worldwide, is implemented on top of an infrastructure of tens of thousands of servers and network components located in many datacenters around the world. At this scale, small and large components fail continuously and the way persistent state is managed in the face of these failures drives the reliability and scalability of the software systems.

    This paper presents the design and implementation of Dynamo, a highly available key-value storage system that some of Amazon’s core services use to provide an “always-on” experience. To achieve this level of availability, Dynamo sacrifices consistency under certain failure scenarios. It makes extensive use of object versioning and application-assisted conflict resolution in a manner that provides a novel interface for developers to use.

    See publication

Patents

  • Dynamic Resource Commitment Management.

    Issued US 8,479,211

    This Elastic Block Storage (EBS) patent describes the management of the block storage to ensure that resources requested by clients are properly provisioned and available. For example, a client can request a resource. The EBS control module can create a volume, specifying the number of partitions and/or the way in which data is stored across partitions to guarantee the requested resource to the client. In addition, partitions and data storage can be updated dynamically without significantly…

    This Elastic Block Storage (EBS) patent describes the management of the block storage to ensure that resources requested by clients are properly provisioned and available. For example, a client can request a resource. The EBS control module can create a volume, specifying the number of partitions and/or the way in which data is stored across partitions to guarantee the requested resource to the client. In addition, partitions and data storage can be updated dynamically without significantly impacting the customer using the volume. The management and update of resources are optimized using various techniques, including striping data, splitting mapping data, and balancing the data in partitions.

  • Managing Route Selection in a Communication Network

    Issued US 8,472,324

    This patent relates to a way to reduce processing load on network routers in our AWS data centers. In particular, the described technology can offload pre-determined route calculations and storage from edge routers to a central processing system in order to increase edge router performance. In some cases, this facilitates the use of lower-capability, cheap commodity-based routers.

  • RESOURCE ISOLATION THROUGH REINFORCEMENT LEARNING

    Issued US US 8429096 B1

    Systems and methods for providing resource isolation in a shared computing environment using reinforcement learning (RL) techniques are disclosed. A resource isolation mechanism may be applied in a shared storage system, or database service, that limits the resource utilization of each namespace to its specified allocation. For example, the resource isolation mechanism may be used to limit the I/O utilization of database applications in a shared computing system (e.g., a system supporting a…

    Systems and methods for providing resource isolation in a shared computing environment using reinforcement learning (RL) techniques are disclosed. A resource isolation mechanism may be applied in a shared storage system, or database service, that limits the resource utilization of each namespace to its specified allocation. For example, the resource isolation mechanism may be used to limit the I/O utilization of database applications in a shared computing system (e.g., a system supporting a database service) to a specified limit. In such embodiments, RL techniques may be applied to the system to automatically control the rate of queries made by an application. RL techniques, such as those based on the State-Action-Reward-State-Action (SARSA) method may be effective in controlling the I/O utilization of database applications for different workloads. RL techniques may be applied globally by the service, or may be applied to particular subscribers, applications, shared resources, namespaces, or query types.

    See patent
  • Managing content delivery network service providers

    Issued US 12/272,699

    A system, method, and computer readable medium for managing CDN service providers are provided. A network storage provider storing one or more resources on behalf of a content provider obtains client computing device requests for content. The network storage provider processes the client computing device requests and determines whether a subsequent request for the resource should be directed to a CDN service provider as a function of the updated or processed by the network storage provider…

    A system, method, and computer readable medium for managing CDN service providers are provided. A network storage provider storing one or more resources on behalf of a content provider obtains client computing device requests for content. The network storage provider processes the client computing device requests and determines whether a subsequent request for the resource should be directed to a CDN service provider as a function of the updated or processed by the network storage provider storage component.

    See patent
  • System-aware resource scheduling

    US 8,347,302

More activity by Swami

View Swami’s full profile

  • See who you know in common
  • Get introduced
  • Contact Swami directly
Join to view full profile

People also viewed

Explore collaborative articles

We’re unlocking community knowledge in a new way. Experts add insights directly into each article, started with the help of AI.

Explore More

Add new skills with these courses