1. A method comprising:
instructing by a base node of a coalesced system to at least one other node of the coalesced system to start a process related to private memory that is firmware, each of the base node and the at least one other node including one or more processors, memory, and inputoutput (IO) modules, the process relating to updating the firmware of the at least one other node;
starting the process by each of the at least one other node on each of the at least one other node, the process accessing private memory that is firmware on each of the at least one other node; and,
reporting back results of the process by each of the at least one other node to the base node,
wherein the method is performed without rebooting of any of the at least one other node, such that the firmware of the at least one other node is updated without rebooting of any of the at least one other node.
2. The method of claim 1, further initially comprising receiving a request from a user to perform system-wide functionality related to private memory that is firmware, the process related to private memory that is firmware based on the system-wide functionality requested.
3. The method of claim 1, further initially comprising starting the process by the base node on the base node, the process accessing private memory that is firmware of the base node.
4. The method of claim 1, further comprising prior to starting the process by each of the at least one other node, receiving instruction from the base node by each of the at least one other node to start the process related to private memory that is firmware.
5. The method of claim 1, further comprising receiving the results of the process from each of the at least one other node by the base node.
6. The method of claim 5, further comprising reporting the results received back from each of the at least one other node by the base node to a user having previously requested system-wide functionality related to private memory that is firmware be performed, the process related to private memory that is firmware based on the system-wide functionality requested.
7. The method of claim 1, wherein instructing by the base node to the at least one other node to start the process related to private memory that is firmware comprises one of multicasting a start-process message by the base node to the at least one other node and unicasting the start-process message by the base node to each of the at least one other node.
8. The method of claim 1, wherein instructing by the base node to the at least one other node to start the process related to private memory that is firmware comprises creating a parent process by the base node at the base node and creating a child process by the parent process at each of the at least one other node.
9. A system comprising:
a plurality of nodes coalesced to function as the system and divided into a base node allowing direct user interaction therewith and other nodes allowing indirect user interaction therewith through the base node, each node including one or more processors, memory, and inputoutput (IO) modules; and,
private memory that is firmware at each of the plurality of nodes, the private memory that is firmware of a node accessible only by the node,
the base node governing performance by each of the plurality of nodes of a process related to private memory that is firmware of each of the plurality of nodes to effectuate user-initiated, system-wide performance of functionality related to private memory that is firmware, the process related to private memory that is firmware based on the functionality initiated and relating to updating the firmware of each of the plurality of nodes, the functionality not resulting in rebooting of any of the plurality of nodes, such that the firmware of each of the plurality of nodes is updated without rebooting of any of the plurality of nodes.
10. The system of claim 9, further comprising a network to which each of the plurality of nodes is connected such that the network assists coalescing of the plurality of nodes as the system.
11. The system of claim 9, further comprising at least one of a keyboard and a display connected to the base node to assist direct user interaction with the base node.
12. The system of claim 9, wherein at least one of the plurality of nodes each comprise a non-computer device having a processor in addition to having the private memory that is firmware.
13. The system of claim 9, wherein the base node governs the performance by each of the plurality of nodes of the process to effectuate the user-initiated, system-wide performance of functionality related to private memory that is firmware by employing a multi-processor protocol (MPP).
14. The system of claim 13, wherein the MPP is part of one of an operating system (OS) of the base node and boot-time services of the base node.
15. An article comprising:
a computer-readable medium; and,
means in the medium for governing performance by each of a plurality of nodes coalesced to function as a system of a process related to private memory that is firmware of each of the plurality of nodes, to effectuate user-initiated, system-wide performance of functionality related to private memory that is firmware, the process related to private memory that is firmware based on the functionality initiated and related to updating the firmware of each of the plurality of nodes, the functionality not resulting in rebooting of any of the plurality of nodes, such that the firmware of the plurality of nodes is updated without rebooting of any of the plurality of nodes,
wherein each of the plurality of nodes includes one or more processors, memory, and inputoutput (IO) modules.
16. The article of claim 15, wherein the medium is a recordable data storage medium.
The claims below are in addition to those above.
All refrences to claim(s) which appear below refer to the numbering after this setence.
1. A best effort client rate limiter apparatus, comprising:
an average rate monitor, the average rate monitor comprising a counter configured to regulate a count value based on a transaction occurrence between an arbiter configured to grant access to a shared resource and one of a plurality of normal clients, the average rate monitor being configured to output a control signal based on the count value of the counter; and
a flow controller configured to regulate a flow of transaction messages between the arbiter and a best effort client, the best effort client having a lower priority access to the shared resource than the plurality of normal clients;
the flow controller being further configured to selectively prevent the flow of the transaction messages between the best effort client and the arbiter based on the control signal, the control signal including an indication that the count value of the counter exceeds a predetermined threshold.
2. The rate limiter as set forth in claim 1, wherein:
the counter is configured to increase the count value in response to an indication that a transaction occurred between the arbiter and one of the plurality of normal clients.
3. The rate limiter as set forth in claim 1, wherein the counter of the average rate monitor is configured to increase or decrease the count value in response to a presence or an absence of an indication that a transaction occurred between the arbiter and one of the plurality of clients.
4. The rate limiter as set forth in claim 3, wherein the counter is configured to decrease by 1 per clock cycle in which the transaction occurrence indication is absent.
5. The rate limiter as set forth in claim 2, wherein the count value of the counter is decremented by an amount of 1, in response to an absence of an indication that a transaction occurred between the arbiter and one of the plurality of clients, until the count value reaches the predetermined threshold value of zero.
6. The rate limiter as set forth in claim 2, wherein the count value of the counter is increased by an amount x\u22121, in response to a presence of an indication that a transaction occurred between the arbiter and one of the plurality of clients, where x is a minimal average period between start times of a first transaction and a second transaction.
7. The rate limiter as set forth in claim 1, wherein the flow controller comprises an AND gate and is configured to receive the control signal of the average rate monitor as a first input, receive a best effort client request line as a second input, and control the flow of the transaction messages between the best-effort client and the arbiter.
8. The rate limiter as set forth in claim 1, wherein the flow controller comprises an AND gate and is configured to receive the control signal of the average rate monitor as a first input and to receive an arbiter grant line as a second input, and control the flow of the transaction messages between the best-effort client and the arbiter.
9. A network device, comprising:
at least one normal service port and at least one best effort port communicating respective packets to a packet processor governed by an arbiter; wherein the packet processor has a predetermined maximum input packet rate for receiving packets;
a packet rate determination module configured to determine whether communication of one or more best effort packets from the best effort port would cause a rate of packets supplied to the packet processor to exceed the maximum input packet rate, and
a rate limiter disposed between the best effort port and the arbiter, the rate limiter being configured to rate limit the communication of packets from the best effort port to the packet processor based on a determination of the packet rate determination module that the supply of best effort packets to the packet processor would cause the packet processor to exceed a maximum input packet rate.
10. The network device as set forth in claim 9, wherein the rate limiter further comprises:
an average rate monitor configured to output a control signal to control the flow controller based on an occurrence of a transaction; and
a flow controller configured to communicate transaction messages between the best effort port and the arbiter;
the flow controller being further configured to prevent the communication of a given best effort port packet by preventing the transaction messages from being communicated between the best effort port and the arbiter when a count of a counter of the rate limiter exceeds a predetermined threshold.
11. The network device as set forth in claim 10, wherein:
the average rate monitor further comprises a counter which increases or decreases in response to the absence of presence of a transaction occurrence indication; and
the average rate monitor being configured to make the determination that the communication of the given best effort port packet would result in exceeding the predetermined maximum input packet rate when the counter exceeds the predetermined threshold.
12. The network device as set forth in claim 11, wherein the counter of the average rate monitor is configured to decrease in response to an absence of a transaction occurrence indication and increase in response to a presence of a transaction occurrence.
13. The network device as set forth in claim 12, wherein the counter is configured to decrease by 1 per clock cycle in which there is an absence of a transaction occurrence indication.
14. The network device as set forth in claim 11, further comprising a balance counter which is configured to decrement by an amount of 1 to a threshold number having a positive value.
15. The network device as set forth in claim 11, wherein the counter is configured to be increase by the average rate monitor, in response to the transaction occurrence indication by an amount x\u22121, where x is a minimal average period between transaction start times.
16. The network device as set forth in claim 12, wherein the flow controller comprises an AND gate and is configured to receive the control signal of the average rate monitor as a first input and a best effort port request line as a second input.
17. The network device as set forth in claim 12, wherein the flow controller comprises an AND gate and is configured to receive the control signal of the average rate monitor as a first input and an arbiter grant line as a second input.
18. The network device as set forth in claim 10, further comprising a second best effort port, wherein:
the flow controller is configured to communicate transaction messages between the second best effort port and the arbiter; and
the flow controller is configured to prevent the transaction messages from being communicated between the second best effort port and the arbiter based on the control signal.
19. A network device comprising:
a flow controller configured to transmit transaction messages between a best effort client and an arbiter, and
an average rate monitor which comprises a counter configured to receive transaction occurrences from the arbiter and count up at an amount of (x\u22121) in response to an indication of the presence of a transaction occurrence, and count down at an amount of 1 per clock cycle, in response to an absence of a transaction occurrence,
wherein x is a minimal average period between transaction start times, and
wherein the arbiter grants a request by the best-effort client to receive a resource when the counter has a value of zero.
20. A method of controlling the flow of packets from a best effort client to a packet processor, the method comprising:
receiving, by a flow controller, transaction messages that are sent between the best effort client and an arbiter;
receiving, by an average rate monitor, transaction occurrences from the arbiter and incrementing a counter by an amount of (x\u22121) if a transaction occurrence is detected, and decrementing the counter at an amount of 1 per clock cycle, if a transaction occurrence is not detected,
wherein \u201cx\u201d is a minimal average period between transaction start times, and
wherein packets are transmitted from the best-effort client to the packet processor when the counter has a value of zero.