| Unexpected bursty traffic due to certain sudden events,such as news in the spotlight on a social network or discounted items on sale,can cause severe load imbalance in backend services.Migrating hot data — the standard approach to achieve load balance — meets a challenge when handling such unexpected load imbalance,because migrating data will slow down the server that is already under heavy pressure.This paper proposes Post Man based on network function virtualization,an emerging paradigm,by which network services can be flexibly provisioned as virtual network functions(VNFs)deployed on general servers.Instead of data migration,Post Man acts as an alternative approach to rapidly mitigate load imbalance for services processing small requests.Motivated by the observation that processing large packets incurs far less CPU overhead than processing small ones,Post Man deploys a number of middleboxes called helpers to assemble small packets into large ones for the heavily-loaded server.This approach essentially offloads the overhead of packet processing from the heavily-loaded server to others.In order to make an illusion that there are direct connections between clients and servers,we design a library at both the server and client side to achieve transparency.With the characteristics of NFV,Post Man can be easily migrated among servers to support scalability and load balancing.To minimize the overhead,Post Man activates helpers on demand,only when bursty traffic is detected.To tolerate helper failures,Post Man can migrate connections across helpers and can ensure packet ordering despite such migration.Our evaluation shows that,with the help of Post Man,a Memcached and Paxos server can mitigate bursty traffic within hundreds of milliseconds,while migrating data takes tens of seconds and increases the latency during migration.Further investigation shows that this is because Post Man can improve the goodputs of Memcached and Paxos by 3.3×and 2.8×,respectively. |
PDF Full Text Request