Estimating cardinality distributions in network traffic
View Patent ↗In one embodiment, a method of monitoring a network. The method includes: receiving, from each host of a set of two or more hosts of the network, a corresponding vector of M components constructed based on data packets received at the host during a time period, M being an integer greater than 1; and, based on the constructed vectors, using an expectation-maximization algorithm to estimate a cardinality distribution for the hosts in the set, wherein constructing a vector includes updating a component of the vector of the corresponding host in response to the corresponding host receiving a data packet, the updating including selecting the component for updating by hashing one or more fields of the data packet received by the corresponding host.
1. A method of monitoring a network, the method comprising:
receiving, from each host of a set of two or more hosts of the network, a corresponding vector of M components constructed based on data packets received at the host during a time period, M being an integer greater than 1; and
based on the constructed vectors, using an expectation-maximization algorithm to estimate a cardinality distribution for the hosts in the set, wherein constructing a vector includes updating a component of the vector of the corresponding host in response to the corresponding host receiving a data packet, the updating including selecting the component for updating by hashing one or more fields of the data packet received by the corresponding host.
2. The method of claim 1 , wherein the estimate of the cardinality distribution for the hosts in the set is an estimate of the cardinality distribution of the number of hosts in the set that communicate with a given number or range of numbers of peers.
3. The method of claim 1 , wherein the estimate of the cardinality distribution for the hosts in the set is an estimate of the cardinality distribution of the number of hosts in the set that have a given number or range of numbers of flows.
4. The method of claim 1 , wherein the updating further comprises determining a number to assign to the component for updating based on the fields of the data packet received by the one of the hosts.
5. The method of claim 4 , wherein constructing a vector involves updating the number assigned to each component of one of the vectors by a process that changes the assigned number in a monotonic manner.
6. The method of claim 1 , wherein the estimating involves evaluating cardinality distribution shared by the vectors.
7. A server comprising:
a processor adapted to:
receive, from each host of a set of two or more hosts of a network, a corresponding vector of M components constructed based on data packets received at the host during a time period, M being an integer greater than 1; and
based on the constructed vectors, use an expectation-maximization algorithm to estimate a cardinality distribution for the hosts in the set, wherein constructing a vector includes updating a component of the vector of the corresponding host in response to the corresponding host receiving a data packet, the updating including selecting the component for updating by hashing fields of the data packet received by the corresponding host.
8. The server of claim 7 , wherein the estimate of the cardinality distribution for the hosts in the set is an estimate of the cardinality distribution of the number of hosts in the set that communicate with a given number or range of numbers of peers.
9. The server of claim 7 , wherein the estimate of the cardinality distribution for the hosts in the set is an estimate of the cardinality distribution of the number of hosts in the set that have a given number or range of numbers of flows.
10. The server of claim 7 , wherein the updating further comprises determining a number to assign to the component for updating based on the one or more fields of the data packet received by the one of the hosts.
11. The server of claim 10 , wherein constructing a vector of M components involves updating the number assigned to each component of one of the vectors by a process that changes the assigned number in a monotonic manner.
12. The server of claim 7 , wherein the estimate involves evaluating cardinality distribution shared by the vectors.
13. A network comprising:
a set of two or more hosts, each host adapted to construct a corresponding vector of M components based on data packets received at the host during a time period, M being an integer greater than 1; and
a server adapted, based on the constructed vectors, to use an expectation-maximization algorithm to estimate a cardinality distribution for the hosts in the set, wherein constructing a vector includes updating a component of the vector of the corresponding host in response to the corresponding host receiving a data packet, the updating including selecting the component for updating by hashing one or more fields of the data packet received by the corresponding host.
14. The network of claim 13 , wherein the estimate of the cardinality distribution for the hosts in the set is an estimate of the cardinality distribution of the number of hosts in the set that communicate with a given number or range of numbers of peers.
15. The network of claim 13 , wherein the estimate of the cardinality distribution for the hosts in the set is an estimate of the cardinality distribution of the number of hosts in the set that have a given number or range of numbers of flows.
16. The network of claim 13 , wherein the updating further comprises determining a number to assign to the component for updating based on the one or more fields of the data packet received by the one of the hosts.
17. The network of claim 16 , wherein constructing a vector of M components involves updating the number assigned to each component of one of the vectors by a process that changes the assigned number in a monotonic manner.
18. The network of claim 13 , wherein the estimate involves evaluating cardinality distribution shared by the vectors.