Erasure coding for data storage

1,294 views · Published 13 April 2016 · 1:26:00 · Indexed 25 September 2026

Channel: ФКН ВШЭ · 2016 · Education

Watch on YouTube

Historically, most large distributed storage systems (e.g., Hotmail) have been using replication to provide reliability against machine failures. Today however as the amount of stored data reaches multiple Zettabytes keeping few copies of data around is prohibitively expensive. Therefore more and more systems are adopting erasure coding in place of replication. Coding for distributed storage has some unique challenges and trade-offs. In this talk, we review the state of the art in erasure codes for distributed storage.

Speaker: Sergey Yekhanin, Microsoft Research.

Workshop on Theoretical Computer Science 2016: https://cs.hse.ru/en/big-data/tcs-lab/tcs2016/
Faculty of Computer Science: https://cs.hse.ru/en/
Follow us: https://www.facebook.com/hsefcs, https://twitter.com/CS_HSE

More from this channel

• 1:09:43 [Коллоквиум]: Resourse Allocation in the Cloud - Assaf Schuster, Techion

  ФКН ВШЭ

  203 views · 9 December 2014

• 1:17:41 [Коллоквиум]: Probabilistic graphical models: Factor graphs and more

  ФКН ВШЭ

  3,453 views · 3 January 2015

• 48:35 [ЗШ 2015]: Как оценить спрос на высшее образование?

  ФКН ВШЭ

  240 views · 11 February 2015

• 1:02:29 [ИТ-лекторий] Большие данные в спортивной индустрии - SAP

  ФКН ВШЭ

  989 views · 20 February 2015

• 50:25 [ДДШ]: Work is fun

  ФКН ВШЭ

  2,079 views · 22 April 2015

• 1:24:40 Testing sparse graph properties

  ФКН ВШЭ

  318 views · 13 April 2016

• 1:28:09 [Коллоквиум]: Physics Informed Machine Learning

  ФКН ВШЭ

  931 views · 19 April 2016

• 48:20 [ДДШ-2016]: Программное обеспечение: от микроконтроллеров до облачных вычислений

  ФКН ВШЭ

  379 views · 21 April 2016