BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Computer Science and Engineering - ECPv6.13.0//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-WR-CALNAME:Computer Science and Engineering
X-ORIGINAL-URL:https://homecse.iitd.ac.in
X-WR-CALDESC:Events for Computer Science and Engineering
REFRESH-INTERVAL;VALUE=DURATION:PT1H
X-Robots-Tag:noindex
X-PUBLISHED-TTL:PT1H
BEGIN:VTIMEZONE
TZID:Asia/Kolkata
BEGIN:STANDARD
TZOFFSETFROM:+0530
TZOFFSETTO:+0530
TZNAME:IST
DTSTART:20240101T000000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;VALUE=DATE:20241202
DTEND;VALUE=DATE:20241208
DTSTAMP:20260927T154511
CREATED:20241126T001433Z
LAST-MODIFIED:20241126T001433Z
UID:182-1733117400-1733549399@homecse.iitd.ac.in
SUMMARY:Winter School 2024 on Data Systems
DESCRIPTION:The Winter School 2024 on Data Systems\, organized by the Data Systems Group of the Department of Computer Science and Engineering at IIT Delhi\, will be held from December 2nd to December 6th\, 2024. Supported by the Mohit Aron Endowment\, this winter school provides an exceptional opportunity for final-year undergraduate students\, master’s and PhD students\, and industry professionals to deepen their knowledge in scalable systems for Big Data\, Data Science and AI. \nParticipants will engage in lectures and hands-on lab sessions on a range of data systems topics\, led by Prof. Kaustubh Beedkar and Prof. Abhilash Jindal. \nRegistration\nApplications are open to final-year undergraduate students\, master’s and PhD students\, and industry professionals with an interest in scalable systems for Big Data\, Data Science\, and AI. The application process for Winter School 2024 consists of two rounds: \nFirst Round – Application Submission\nInterested candidates must submit their applications via the provided online form by 5:00 pm on 14th November 2024. Only candidates who complete this submission will be considered for the next round. \nSecond Round – Online Test\nCandidates who have successfully applied in the first round will be invited to participate in an online test scheduled for 16th November 2024 17th November 2024 at 10:00 AM. \nSelected candidates will receive notification of their acceptance into the Winter School by 17th November 2024 21st November 2024. \nLogistics\nVenue: Department of Computer Science and Engineering\, IIT Delhi\nDates: December 2nd – December 6th\, 2024\nAccommodation: Hostel accommodation including breakfast\, lunch\, dinner\, and high tea will be provided from Dec 1st – Dec 7th\, 2024. In addition\, a stipend of 2000 INR will be provided to selected students. Participants will be required to check in on December 1st\, 2024\, and check out before noon on December 7th\, 2024.\nAfter winter school: Participants who successfully complete the Winter School will receive a Certificate of Participation. Please note that we will not provide recommendation letters for graduate school applications. Students will be encouraged to apply for paid internships at IIT Delhi during the 8 weeks of summer 2025.\nDaily Schedule\n09:00 am – 10:15 am: Lecture\n10:15 am – 10:45 am: High tea + offline discussions\n10:45 am – 01:00 pm: Lectures with breaks\n01:00 pm – 02:00 pm: Lunch Break\n02:00 pm – 05:00 pm: Lab\n03:30 pm – 04:00 pm: High tea \nProgram\nThe Winter School will cover the following topics: \nDay 1: \n\nDistributed data processing\nHoly grail: transparently scale\, and tolerate faults/stragglers;\nDifficulties in transparent locality\, scalability\, fault tolerance\, and straggler mitigation;\nIntroduction to checkpointing and replication;\nIntroduction to distributed data processing with MapReduce;\nMain ideas: functional programming model; separate control and data plane; locality optimizations\, re-execute lost/backup tasks\, deterministic/idempotent tasks.\nLab: Write a fault tolerant computation from scratch\n\nDay 2: \n\nDataflow systems for batch processing\nKeynote address: TBA\nSpark’s Resilient distributed dataset (RDD) abstraction: write-once for consistent replication\, coarse-grained transformations to reduce lineage\, lineage-based re-execution of lost/backup tasks;\nAbstractions beyond MapReduce\, lazy execution\, stage planning\, narrow and wide-dependencies\, query optimizations.\nLab: Hands-on exercises with Spark.\n\nDay 3: \n\nDataflow systems for stream processing\nSemantics of stream processing: unbounded streams\, event time vs processing time\, hopping/sliding/session windows\, watermarks;\nStreaming query optimizations and continuous operator model of Flink for low-latency;\nFault tolerance: problems due to state in continuous operators\, pull state out of operators by discretizing streams\, consistent checkpoints in Flink using Chandy-Lamport algorithm.\nLab: Hands-on exercises with Flink.\n\nDay 4: \n\nCyclic dataflow systems\nIterative data processing: solving graph problems with iterative data processing\, driver-based iterations vs native iterations with cyclic dataflows\, changes for creating consistent checkpoints.\nIntroduction to ML workflows: error handling and correction as a first class citizen\, Reactive dataflow with deletes\, edits\, and appends.\nBackward and forward lineage for backward tracing and delete propagation\, stage planning with incremental/non-incremental\, monotonic/non-monotonic operators\nLab: Hands-on exercises with Popper.\n\nDay 5: \n\nCross-platform data processing\nIntroduction to Wayang: a platform for cross-framework data processing\, allowing optimizations across Spark\, Flink\, and Relational Databases.\nSessions about data systems research at IITD\, guiding participants interested in pursuing higher studies\, and research success stories.\nLab: Hands-on exercises with Wayang.\n\nDisclaimer: The schedule and topics are only tentative. They are subject to change depending upon attendee interests.
URL:https://homecse.iitd.ac.in/event/winter-school-2024-on-data-systems/
LOCATION:Bharti 501\, IIT Campus\, Hauz Khas\, New Delhi
CATEGORIES:Schools/Workshops
END:VEVENT
END:VCALENDAR