TY - GEN
T1 - Dapper
T2 - 2017 Symposium on SDN Research, SOSR 2017
AU - Ghasemi, Mojgan
AU - Benson, Theophilus
AU - Rexford, Jennifer L.
N1 - Publisher Copyright:
© 2017 ACM.
PY - 2017/4/3
Y1 - 2017/4/3
N2 - With more applications moving to the cloud, cloud providers need to diagnose performance problems in a timely manner. Offline processing of logs is slow and inefficient, and instrumenting the end-host network stack would violate the tenants' rights to manage their own virtual machines (VMs). Instead, our system Dapper analyzes TCP performance in real time near the end-hosts (e.g., at the hypervisor, NIC, or top-of-rack switch). Dapper determines whether a connection is limited by the sender (e.g., a slow server competing for shared resources), the network (e.g., congestion), or the receiver (e.g., small receive buffer). Emerging edge devices now offer flexible packet processing at high speed on commodity hardware, making it possible to monitor TCP performance in the data plane, at line rate. We use P4 to prototype Dapper and evaluate our design on real and synthetic traffic. To reduce the data-plane state requirements, we perform lightweight detection for all connections, followed by heavier-weight diagnosis just for the troubled connections.
AB - With more applications moving to the cloud, cloud providers need to diagnose performance problems in a timely manner. Offline processing of logs is slow and inefficient, and instrumenting the end-host network stack would violate the tenants' rights to manage their own virtual machines (VMs). Instead, our system Dapper analyzes TCP performance in real time near the end-hosts (e.g., at the hypervisor, NIC, or top-of-rack switch). Dapper determines whether a connection is limited by the sender (e.g., a slow server competing for shared resources), the network (e.g., congestion), or the receiver (e.g., small receive buffer). Emerging edge devices now offer flexible packet processing at high speed on commodity hardware, making it possible to monitor TCP performance in the data plane, at line rate. We use P4 to prototype Dapper and evaluate our design on real and synthetic traffic. To reduce the data-plane state requirements, we perform lightweight detection for all connections, followed by heavier-weight diagnosis just for the troubled connections.
KW - Measurement
KW - Network monitoring
KW - Performance diagnosis
UR - http://www.scopus.com/inward/record.url?scp=85018964756&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85018964756&partnerID=8YFLogxK
U2 - 10.1145/3050220.3050228
DO - 10.1145/3050220.3050228
M3 - Conference contribution
AN - SCOPUS:85018964756
T3 - SOSR 2017 - Proceedings of the 2017 Symposium on SDN Research
SP - 61
EP - 74
BT - SOSR 2017 - Proceedings of the 2017 Symposium on SDN Research
PB - Association for Computing Machinery, Inc
Y2 - 3 April 2017 through 4 April 2017
ER -