Synchronizing multicast data distribution on a computing device
View Patent ↗A computing device that is configured for synchronizing multicast data distribution is described. The computing device includes a processor and instructions stored in memory. The computing device reads a block of data and sends the block of data via multicast. It is determined whether a packet in the block of data requires a new mapped file view. This determination is based on the capacity of a current mapped file view on a target node. The computing device delays sending more data if the new mapped file view is required.
1. A computing device that is configured for synchronizing multicast data distribution, comprising:
a processor;
memory in electronic communication with the processor;
instructions stored in the memory, the instructions being executable to:
read a first block of data;
send the first block of data via multicast, wherein the first block of data comprises software files for an application or program;
send a message to a target node that specifies a capacity of a current mapped file view for the target node, wherein the current mapped file view comprises a first portion of the memory that is mapped to a first portion of computing device storage corresponding to a first portion of a file, wherein the file is one of the software files;
determine whether the first block of data requires a new mapped file view by determining the difference between the capacity of the current mapped file view sent in the message and the size of the first block of data, wherein the new mapped file view is required when the size of the first block of data exceeds the capacity of the current mapped file view, and wherein the new mapped file view comprises a second portion of the memory that is mapped to a second portion of computing device storage corresponding to a second portion of the file, wherein the file is one of the software files; and
delay sending a second block of data based on whether the first block of data requires a new mapped file view.
2. The computing device of claim 1 , wherein delaying sending the second block of data allows one or more target nodes to map a new mapped file view.
3. The computing device of claim 1 , wherein delaying sending a second block of data comprises suspending sending the second block of data.
4. The computing device of claim 1 , wherein delaying sending a second block of data comprises maintaining a network bandwidth usage in a certain range.
5. The computing device of claim 1 , wherein the capacity of the current mapped file view is based on a number of packets.
6. The computing device of claim 1 , wherein the capacity of the current mapped file view is determined by measuring past system performance.
7. The computing device of claim 6 , wherein the past system performance is measured by a measurement selected from the group consisting of: a number of retransmission requests, a number of dropped packets and a time used to complete a multicast operation.
8. The computing device of claim 1 , wherein the instructions are further executable to:
receive a message from a target node, wherein the message specifies a processor speed, a memory write speed and a total memory capacity from the target node; and
determine the capacity of the current mapped file view based on the processor speed, memory write speed and total memory capacity.
9. The computing device of claim 8 , wherein the instructions are further executable to:
receive the message from a plurality of target nodes; and
determine a delay based on the determined capacity of the current mapped file view and a slowest target node, wherein the slowest target node is the target node that requires the longest amount of time to map the new mapped file view.
10. The computing device of claim 1 , wherein the instructions are further executable to receive a message from a target node requesting a specified delay and wherein delaying sending the second block of data is based on the specified delay.
11. The computing device of claim 1 , wherein sending the second block of data is delayed until a message is received from a target node indicating that the target node has finished mapping the new mapped file view.
12. A method for synchronizing multicast data distribution on a computing device, comprising:
reading a first block of data;
sending the first block of data via multicast from the computing device, wherein the first block of data comprises software files for an application or program;
sending a message to a target node that specifies a capacity of a current mapped file view for the target node, wherein the current mapped file view comprises a first portion of the memory that is mapped to a first portion of computing device storage corresponding to a first portion of a file, wherein the file is one of the software files;
determining, on the computing device, whether the first block of data requires a new mapped file view, by determining the difference between the capacity of the current mapped file view sent in the message and the size of the first block of data, wherein the new mapped file view is required when the size of the first block of data exceeds the capacity of the current mapped file view, and wherein the new mapped file view comprises a second portion of a memory that is mapped to a second portion of computing device storage corresponding to a second portion of the file, wherein the file is one of the software files; and
delaying sending a second block of data from the computing device based on whether the first block of data requires a new mapped file view.
13. The method of claim 12 , wherein delaying sending the second block of data allows one or more target nodes to map a new mapped file view.
14. The method of claim 12 , wherein delaying sending a second block of data comprises suspending sending the second block of data.
15. The method of claim 12 , wherein delaying sending a second block of data comprises maintaining a network bandwidth usage in a certain range.
16. The method of claim 12 , wherein the capacity of the current mapped file view is based on a number of packets.
17. The method of claim 12 , wherein the capacity of the current mapped file view is determined by measuring past system performance.
18. The method of claim 17 , wherein the past system performance is measured by a measurement selected from the group consisting of: a number of retransmission requests, a number of dropped packets and a time used to complete a multicast operation.
19. The method of claim 12 , further comprising:
receiving a message from a target node, wherein the message specifies a processor speed, a memory write speed and a total memory capacity from the target node; and
determining the capacity of the current mapped file view based on the processor speed, memory write speed and total memory capacity.
20. The method of claim 19 , further comprising:
receiving the message from a plurality of target nodes; and
determining a delay based on the determined capacity of the current mapped file view and a slowest target node, wherein the slowest target node is the target node that requires the longest amount of time to map the new mapped file view.
21. The method of claim 12 , further comprising receiving a message from a target node requesting a specified delay and wherein delaying sending the second block of data is based on the specified delay.
22. The method of claim 12 , wherein sending the second block of data is delayed until the computing device receives a message from a target node indicating that the target node has finished mapping the new mapped file view.
23. A non-transitory computer-readable medium for synchronizing multicast data distribution on a computing device comprising executable instructions for:
reading a first block of data;
sending the first block of data via multicast, wherein the first block of data comprises software files for an application or program;
sending a message to a target node that specifies a capacity of a current mapped file view for the target node, wherein the current mapped file view comprises a first portion of the memory that is mapped to a first portion of computing device storage corresponding to a first portion of a file, wherein the file is one of the software files;
determining whether the first block of data requires a new mapped file view, by determining the difference between the capacity of the current mapped file view sent in the message and the size of the first block of data, wherein the new mapped file view is required when the size of the first block of data exceeds the capacity of the current mapped file view, and wherein the new mapped file view comprises a second portion of a memory that is mapped to a second portion of computing device storage corresponding to a second portion of the file, wherein the file is one of the software files; and
delaying sending a second block of data based on whether the first block of data requires a new mapped file view.