Methods and devices for improved percentile extraction of network monitoring data
Described are various embodiments of a device and method for computing statistics of various network monitoring metrics. In one embodiment, the method includes constructing a first histogram of network traffic monitoring data acquired over a designated sampling period from said network; identifying one or more bins of said first histogram comprising each at least one of one or more desired percentile values; for each identified bin: building a second histogram centered on said identified bin, said second histogram comprising a second bin size that is smaller than said first bin size; calculating one or more bins of said second histogram comprising each at least one of said one or more desired percentile values and the values associated therewith; and converting said values associated therewith into percentile values representative of the range defined between said lower order of magnitude to said higher order of magnitude.
1. A computer-implemented network monitoring method, the method comprising:
acquiring, by a processor, network traffic monitoring data over a designated sampling period from said network, the network traffic monitoring data comprising a plurality of packet-related time values extending from a lower order of magnitude to a higher order of magnitude;
constructing, by the processor, a first histogram representative of said network traffic monitoring data comprising a first bin size and covering a range of said lower order of magnitude to said higher order of magnitude;
identifying, by the processor, one or more bins of said first histogram comprising each at least one of one or more desired percentile values of said network traffic monitoring data;
for each identified bin of said one or more bins of said first histogram:
building, by the processor, a second histogram centered on said identified bin, said second histogram comprising a second bin size that is smaller than said first bin size;
calculating, by the processor, one or more bins of said second histogram comprising each at least one of said one or more desired percentile values and the values associated therewith; and
converting, for each of said second histogram, each of said values associated therewith into percentile values representative of the range defined between said lower order of magnitude to said higher order of magnitude; and
wherein said building and said calculating is done recursively, wherein the said calculated one or more bins of said second histogram from a first iteration become the identified one or more bins of said first histogram in a following iteration.
2. The method of claim 1 , wherein said building and said calculating is done recursively until the second bin size of the last iteration is smaller than a threshold.
3. The method of claim 1 , wherein said building and said calculating is done recursively for a predetermined number of iterations.
4. The method of claim 1 , wherein said packet-related time values comprises one of: round trip time, one-way delay or two-way delay.
5. The method of claim 1 , wherein said building and said calculating is done in parallel for each identified bin.
6. The method of claim 1 , wherein said building includes:
subtracting from all packet-related time values the bin number of the identified bin of said first histogram multiplied by the first bin size;
clipping all negative values obtained from said subtraction to a value of zero; and
clipping all values greater than the first bin size obtained from said subtraction to a value of said first bin size.
7. The method of claim 1 , wherein said converting comprises adding to each of said values associated therewith a multiplication of the weight of each identified bin with the corresponding bin number.
8. The method of claim 1 , further comprising the steps of:
transmitting, by the processor, via said network, said one or more desired percentile values to a reporting system.
9. The method of claim 1 , wherein said method is repeated according to a predetermined time interval.
10. A network monitoring device, the device comprising:
one or more processors;
a network module communicatively coupled to said one or more processors and configured to send and receive packets via a network;
a memory storing instructions that, when executed by the one or more processors, configured the device to:
acquire network traffic monitoring data over a designated sampling period from said network, the network traffic monitoring data comprising a plurality of packet-related time values extending from a lower order of magnitude to a higher order of magnitude;
construct a first histogram representative of said network traffic monitoring data comprising a first bin size and covering a range of said lower order of magnitude to said higher order of magnitude;
identify one or more bins of said first histogram comprising each at least one of one or more desired percentile values of said network traffic monitoring data;
for each identified bin of said one or more bins of said first histogram:
build a second histogram centered on said identified bin, said second histogram comprising a second bin size that is smaller than said first bin size;
calculate one or more bins of said second histogram comprising each at least one of said one or more desired percentile values and the values associated therewith; and
convert each of said values associated therewith into percentile values representative of the range defined between said lower order of magnitude to said higher order of magnitude; and
wherein said building and said calculating is done recursively, wherein the said calculated one or more bins of said second histogram from a first iteration become the identified one or more bins of said first histogram in a following iteration.
11. The device of claim 10 wherein said building and said calculating is done recursively until the second bin size of the last iteration is smaller than a threshold.
12. The device of claim 10 wherein said building and said calculating is done recursively for a predetermined number of iterations.
13. The device of claim 10 , wherein said packet-related time values comprises one of: round trip time, one-way delay or two-way delay.
14. The device of claim 10 , wherein sad building and said calculating is done in parallel for each identified bin.
15. The device of claim 10 , wherein said building includes:
subtracting from all packet-related time values the bin number of the identified bin of said first histogram multiplied by the first bin size;
clipping all negative values obtained from said subtraction to a value of zero; and
clipping all values greater than the first bin size obtained from said subtraction to a value of said first bin size.
16. The device of claim 10 , wherein said converting comprises adding to each of said values associated therewith a multiplication of the weight of each identified bin with the corresponding bin number.
17. The device of claim 10 , wherein said instructions further configure the device to:
transmit said one or more desired percentile values to a reporting system via said network.
18. The device of claim 10 , wherein said instructions are repeated according to a predetermined time interval.