Explicit constructions of MSR codes for clustered distributed storage: the rack-aware storage model
From MaRDI portal
(Redirected from Publication:5211640)
Abstract: The paper is devoted to the problem of erasure coding in distributed storage. We consider a model of storage that assumes that nodes are organized into equally sized groups, called racks, that within each group the nodes can communicate freely without taxing the system bandwidth, and that the only information transmission that counts is the one between the racks. This assumption implies that the nodes within each of the racks can collaborate before providing information to the failed node. The main emphasis of the paper is on code construction for this storage model. We present an explicit family of MDS array codes that support recovery of a single failed node from any number of helper racks using the minimum possible amount of inter-rack communication (such codes are said to provide optimal repair). The codes are constructed over finite fields of size comparable to the code length. We also derive a bound on the number of symbols accessed at helper nodes for the purposes of repair, and construct a code family that approaches this bound, while still maintaining the optimal repair property. Finally, we present a construction of scalar Reed-Solomon codes that support optimal repair for the rack-oriented storage model.
Recommendations
Cited in
(16)- Architecture-aware coding for distributed storage: repairable block failure resilient codes
- Johnson graph codes
- A new minimize matrix computation coding method for distributed storage systems
- Functional repair codes: a view from projective geometry
- A new piggybacking design for systematic MDS storage codes
- A new (k+2,k) Hadamard minimum storage regenerating code
- Research on distributed storage problems over unidirectional ring networks
- MDS code constructions with small sub-packetization and near-optimal repair bandwidth
- Multilinear algebra for distributed storage
- Cooperative repair of multiple node failures in distributed storage systems
- Scalar MSCR codes via the product matrix construction
- An exponential lower bound on the sub-packetization of MSR codes
- Codes for Distributed Storage
- A class of minimum storage cooperative regenerating codes with low access property
- MDS array codes with efficient repair and small sub-packetization level
- (1+)-optimal MDS codes: contacting any set of helper nodes smaller than n-1
This page was built for publication: Explicit constructions of MSR codes for clustered distributed storage: the rack-aware storage model
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5211640)
