Simple Load Balancing for Distributed Hash Tables Export

@incollection{citeulike:3407835, abstract = {Distributed hash tables have recently become a useful building block for a variety of distributed applications. However, current schemes based upon consistent hashing require both considerable implementation complexity and substantial storage overhead to achieve desired load balancing goals. We argue in this paper that these goals can be achieved more simply and more cost-effectively. First, we suggest the direct application of the power of two choices paradigm, whereby an item is stored at the less loaded of two (or more) random alternatives. We then consider how associating a small constant number of hash values with a key can naturally be extended to support other load balancing strategies, including load-stealing or load-shedding, as well as providing natural fault-tolerance mechanisms.}, author = {Byers, John and Considine\$^⋆\$, Jeffrey and Mitzenmacher, Michael}, citeulike-article-id = {3407835}, citeulike-linkout-0 = {http://www.springerlink.com/content/r9r4qcqxc2bmfqmr}, journal = {Peer-to-Peer Systems II}, keywords = {2003, bins-and-balls, dht, distribution, load\_balancing, p2p}, pages = {80--87}, posted-at = {2009-01-18 20:56:43}, priority = {2}, title = {Simple Load Balancing for Distributed Hash Tables}, url = {http://www.springerlink.com/content/r9r4qcqxc2bmfqmr}, year = {2003} }

TY - CHAP ID - citeulike:3407835 L3 - citeulike-article-id:3407835 N2 - Distributed hash tables have recently become a useful building block for a variety of distributed applications. However, current schemes based upon consistent hashing require both considerable implementation complexity and substantial storage overhead to achieve desired load balancing goals. We argue in this paper that these goals can be achieved more simply and more cost-effectively. First, we suggest the direct application of the power of two choices paradigm, whereby an item is stored at the less loaded of two (or more) random alternatives. We then consider how associating a small constant number of hash values with a key can naturally be extended to support other load balancing strategies, including load-stealing or load-shedding, as well as providing natural fault-tolerance mechanisms. JF - Peer-to-Peer Systems II EP - 87 TI - Simple Load Balancing for Distributed Hash Tables SP - 80 KW - 2003 KW - bins-and-balls KW - dht KW - distribution KW - load_balancing KW - p2p AU - Byers, John AU - Considine$^⋆$, Jeffrey AU - Mitzenmacher, Michael PY - 2003/// UR - http://www.springerlink.com/content/r9r4qcqxc2bmfqmr ER -