User login

Stephen Eichler's blog

22

Oct

2013

Monitored Caida and Planetlab as more warts data is produced. Caida is collecting data with an updated driver for measuring the efficiency of several methods of choosing the flow ID when running MDA traceroute.

Data from the second round of non flow ID field analysis has been processed to raw data and then summarised using a spread sheet. There seem to be a small number routers that can segregate packets to different successor nodes based on a number of non standard fields (not the classic 5-tuple).

The slide set for the PhD conference has been updated. More graphs with error bars for 95% confidence have been produced, including a graph of the percentage of paths with a particular type of load balancer, and this was done for the three usual packet types.

15

Oct

2013

Finished off gathering field testing data on Caida. Began the next run which is testing the efficiency of several flow ID selection modes. It was necessary to update the driver which was previously used to collect this data type on Yoyo.

More work has been done on the slide set for the conference. More statistical tests have been carried out and more graphs have been produced from the raw data. Further pruning of the slide set has been carried out.

08

Oct

2013

I have been monitoring the Caida run of non 5-tuple field analysis. I have started to download the completed warts sets.

A dump of load balancer ID and next hops was included in the data analysis of the warts files from Caida and Planetlab. A perl program using this data was written to count load balancers found by only one of the vantage points. This is to help determine how affective the coverage we have is.

Data from the fourth scamper run has been processed, comparing it to the first run. These results have been incorporated into the conference slides. Furthermore the introduction of the slide set has been extended and more graphs have been added. Some pruning has also been carried out.

The large analysis of Caida runs on wraith is still running. This should provide information about how many new load balancers are added with each new vantage point. This is using quite a lot of memory as the data from all the vantage points is held at the same time, so wraith is ideal for the purpose.

01

Oct

2013

Monitoring and downloading of Ciada scamper warts results has been carried out.

My slides for the conference have been updated in the introduction and in the results from the fields analysis.

A program to output load balancer addresses and next set addresses was written. After this a program to read and process these sets of addresses was written in perl and run on some of the Caida data.

A modification to the fields data processing program was written to check for per packet detection of load balancers also identified as exhibiting per flow behaviour based on a non standard flow field. This program is currently running.

24

Sep

2013

Data collection on Caida and Planetlab has been monitored and data has been downloaded when complete.

Programs have been written to analyse the data from the first half of the 'per flow like fields' collection on Caida. Results have been added to a draft slide set for the students conference.

The Internet simulator has finished its most recent runs.

17

Sep

2013

Some of the warts data files are downloading from the latest scamper run on planetlab.

The scamper warts file analysis methods have been further developed to analyse the successors of load balancing interfaces to find separate diamonds that are based on the same divergence point. As well as this counts have been carried out of destinations found, traces with the maximum limit of packets and both together. In particular analysis of the vantage point with a higher setting of the probing maximum has been carried out.

The Internet Simulator runs have not completed. It may be necessary to check that the limit on the number of traces carried out has worked correctly.

Initial compilation of the slides for the student conference has been carried out. In particular data from the new Caida and Planetlab results has been collected.

10

Sep

2013

The internet simulator has been run using a complete data team from Caida. I was necessary to increase the allowed heap size and the trace list to be carried out was reduced to 100,000.

The next round of scamper data collection on Caida and Planetlab has been initiated.

The modification to the analysis of scamper data has been further updated to count diamonds based on the same load balancing node. There is a certain amount of trial and error involved, so it will take some more time to do this.

03

Sep

2013

Finished downloading the warts files from the last scamper fields run. The fields run collects data about the ability of non load balancing fields to cause load balancing segregation. I have been unable to start the next run so far as some operating systems on the nodes are being upgraded.

The warts churn analysis program has been upgraded to count load balancers with the same divergence point but different successors. Also there has been a count of traces were the destination has been reached.

The internet simulator has been set up for version control and an attempt to use a large data set has been carried out. The simulator seems to have a heap limit of 16GB, so this will need to be increased.

27

Aug

2013

The results from the fields-driver-one scamper run on Caida are being downloaded. This explores non load balancing fields for possible rare load balancing influence.

The results from the scamper run3 have been computed and tabulated in conjunction with runs 1 & 2. These runs measure turnover of load balancers.

A plan for version control of the Internet Simulator has been devised, as the unzipped programs are not active svn copies.

20

Aug

2013

Downloading of warts data from Caida has been carried out.

Analysis of research data has been extended to include the search for load balancer nodes that no longer have more than one successor, but that did on another occasion.

A driver for testing non load balancing fields for load balancing behaviour has been written and tested. The job was divided into two drivers so that there are only six or seven short traces at a time.