Hierarchy-aware distributed overlays in data centers using DC2

Popular online services such as social networks, e-commerce and bidding are routinely hosted in large-scale data centers. Group communication systems (e.g., multicast) and distributed key-value stores are among some of the most essential building blocks for these services. Due to their scaling requirements, overlay networks such as distributed hash tables (DHTs) have been traditionally used in such systems. Modern hierarchical datacenter networks and global services running across datacenters pose unique challenges that traditional systems are ill-equipped to handle. For instance, the inherent multi-rooted tree topology design with oversubscription at the core translates into lesser bandwidth at the upper levels of the trees; traditional systems do not take this into consideration leading to a wastage of precious network resources. To solve this problem, we introduce a hierarchy-aware distributed overlay framework called DC2, for large scale and highly dynamic services. We build two applications-DC2-Multicast and DC2-Store-on top of DC2. In our experiments using a real prototype deployed over 700 nodes running over a Modelnet topology with 2 datacenters, we found that DC2-Multicast minimizes message latencies by several orders of magnitude, and reduces node and link stress by a factor of 2 to 3. We also find a reduction in object lookup latency by a factor of 8.

[1]  Ion Stoica,et al.  SAAR: A Shared Control Plane for Overlay Multicast , 2007, NSDI.

[2]  Scott Shenker,et al.  Internet indirection infrastructure , 2004, IEEE/ACM Transactions on Networking.

[3]  Amin Vahdat,et al.  Scalability in adaptive multi-metric overlays , 2004, 24th International Conference on Distributed Computing Systems, 2004. Proceedings..

[4]  David Mazières,et al.  Kademlia: A Peer-to-Peer Information System Based on the XOR Metric , 2002, IPTPS.

[5]  Amin Vahdat,et al.  Mace: language support for building distributed systems , 2007, PLDI '07.

[6]  Bobby Bhattacharjee,et al.  Scalable application layer multicast , 2002, SIGCOMM '02.

[7]  Dejan Kostic,et al.  Scalability and accuracy in a large-scale network emulator , 2002, CCRV.

[8]  Antony I. T. Rowstron,et al.  Storage management and caching in PAST, a large-scale, persistent peer-to-peer storage utility , 2001, SOSP.

[9]  T. S. Eugene Ng,et al.  The Impact of Virtualization on Network Performance of Amazon EC2 Data Center , 2010, 2010 Proceedings IEEE INFOCOM.

[10]  David R. Karger,et al.  Chord: A scalable peer-to-peer lookup service for internet applications , 2001, SIGCOMM '01.

[11]  Amin Vahdat,et al.  Using Random Subsets to Build Scalable Network Services , 2003, USENIX Symposium on Internet Technologies and Systems.

[12]  John Kubiatowicz,et al.  Handling churn in a DHT , 2004 .

[13]  Antonio F. Gómez-Skarmeta,et al.  Cyclone: a novel design schema for hierarchical DHTs , 2005, Fifth IEEE International Conference on Peer-to-Peer Computing (P2P'05).

[14]  Dan R. K. Ports,et al.  Census: Location-Aware Membership Management for Large-Scale Distributed Systems , 2009, USENIX Annual Technical Conference.

[15]  Brighten Godfrey,et al.  OpenDHT: a public DHT service and its uses , 2005, SIGCOMM '05.

[16]  Albert G. Greenberg,et al.  VL2: a scalable and flexible data center network , 2009, SIGCOMM '09.

[17]  Antony I. T. Rowstron,et al.  Pastry: Scalable, Decentralized Object Location, and Routing for Large-Scale Peer-to-Peer Systems , 2001, Middleware.

[18]  Kirk L. Johnson,et al.  Overcast: reliable multicasting with on overlay network , 2000, OSDI.

[19]  Krishna P. Gummadi,et al.  Canon in G major: designing DHTs with hierarchical structure , 2004, 24th International Conference on Distributed Computing Systems, 2004. Proceedings..

[20]  Miguel Castro,et al.  Scribe: a large-scale and decentralized application-level multicast infrastructure , 2002, IEEE J. Sel. Areas Commun..

[21]  Yoav Tock,et al.  Dr. multicast: Rx for data center communication scalability , 2010, EuroSys '10.

[22]  Amin Vahdat,et al.  Bullet: high bandwidth data dissemination using an overlay mesh , 2003, SOSP '03.

[23]  Werner Vogels,et al.  Dynamo: amazon's highly available key-value store , 2007, SOSP.