Computer System Input/Output Management

Abstract

The collection of performance data at multiple servers in a SAN and forwarding that data to a centralized server for analysis is disclosed. Remote agents and a central server application collect specific interesting negative event data to enable a picture of the operational health of the SAN to be determined. The agents are placed in servers having HBAs acting as initiators. The agents interact with the HBAs through a driver stack to collect event data. Because of the initiator function they perform, HBAs have visibility to parts of the network that other entities do not have access to, and thus are ideal locations for gathering event data. A SAN diagnostics manager then pulls the collected data from each agent so that a “picture” of the SAN can be developed. In addition to collecting initiator data, the agents also collect errors and performance data from the OS of the servers.

1 . A diagnostics manager for a network, the diagnostics manager comprising:
 a collector operable to periodically poll one or more agents of a particular type to collect performance data from the network, wherein each server in the plurality of servers comprises an agent;   a database for storing the performance data; and   an application operable to retrieve the performance data.   
     
     
         2 - 104 . (canceled) 
     
     
         105 . The diagnostics manager of  claim 1 , wherein each server of the plurality of servers comprises an initiator, wherein the initiator is operable to transmit performance data to the agent. 
     
     
         106 . The diagnostics manager of  claim 105 , wherein an input/output (I/O) controller is operable as the initiator. 
     
     
         107 . The diagnostics manager of  claim 106 , wherein an initiator driver stack is operable to communicate with the I/O controller. 
     
     
         108 . The diagnostics manager of  claim 106 , wherein the agent is operable to communicate with the I/O controller through an initiator driver stack to collect initiator event data related to the performance of the network. 
     
     
         109 . The diagnostics manager of  claim 1 , wherein the diagnostics manager comprises a service locations protocol (SLP) interface operable to communicate with the agent to determine an agent type. 
     
     
         110 . The diagnostics manager of  claim 1 , wherein the diagnostics manager is operable to assist in determining a root cause of a network performance issue. 
     
     
         111 . The diagnostics manager of  claim 1 , wherein the network is a storage area network, wherein the agent is a server adapter operable to communicate data and command messages between a server of the plurality of servers and one or more storage systems. 
     
     
         112 . The diagnostics manager of  claim 111 , wherein the server adapter is associated with a server adapter driver, the server adapter driver operable to collect performance data comprising one or more of:
 flow of data and command messages through the server adapter,   the performance data,   performance statistics, and   error statistics.   
     
     
         113 . The diagnostics manager of  claim 1 , wherein the performance data is initiator event data. 
     
     
         114 . A method for managing diagnostics for a network, the method comprising:
 periodically polling one or more agents of a particular type to collect performance data from the network, wherein each server of a plurality of servers comprises an agent;   storing the performance data; and   retrieving the performance data.   
     
     
         115 . The method of  claim 114 , wherein the method comprises transmitting performance data from an initiator to the agent. 
     
     
         116 . The method of  claim 115 , wherein an input/output (I/O) controller is operable as the initiator. 
     
     
         117 . The method of  claim 116 , wherein the method comprises communicating between the agent and the I/O controller through an initiator driver stack. 
     
     
         118 . The method of  claim 116 , wherein the method comprises communicating between the agent and the I/O controller through a service locations protocol (SLP). 
     
     
         119 . The method of  claim 114 , wherein the method comprises communicating with the agent to determine an agent type. 
     
     
         120 . The method of  claim 114 , wherein the method comprises determining a root cause of a network performance issue. 
     
     
         121 . The method of  claim 114 , wherein the network is a storage area network, wherein the method comprises communicating data and command messages between a server of the plurality of servers and one or more storage systems. 
     
     
         122 . The method of  claim 121 , wherein the method comprises collecting performance data concerning the flow of data and command messages, the performance data comprising one or more of:
 flow of data and command messages through the server adapter,   the performance data,   performance statistics, and   error statistics.   
     
     
         123 . The method of  claim 114 , wherein the performance data is initiator event data.