To share what I've learned !

Few years back getting 1Mb bandwidth internet port cost much , the same scenario experienced by the mobile operators. when the competition increases the new price model and service oriented architecture and focus on the data services increases. But the wholesale internet business seems to be facing the same problem . "Wholesale Internet Bandwidth Prices Keep Falling" according to the 2008 report it suggest between 10$ -14$ this values indicate the US pops pricing . But in the Asian countries specially south Asian countries the pricing stricture really high when compared to Europe due to the lack of infrastructure. But the Indian market giving the pressure to reduce the pricing structure. That will have certain impact on the neighbor countries. with this in mind when reaching 2020 the per Gigabit cost will be the same and the services that use to offer will be the dominating factor in the service providers. Hosting may migrate to community offer or hybrid cloud structure. whe...

Some of the ideas I've thought may be some people already done, but need to do literature survey on these areas: 1) Application issue identification based on network analysis. Since i'm working in an ISP sometime really hard to figure out the application issues that affected by network interaction. such as packet loss , throughput , mtu , jitter etc ... analyzing the packet using wireshark or other tools sometimes won't reveal important factors. ( May be i don't know how to use this tool :) ) . initial search yields some commercial tools .... 2) Data management for Network Management data. This is related to my previous post. Data capturing and logging almost standardized but how we are analyzing and how to take action mostly depend on the requirement on hand. Based on the requirements there can be different ideas. * Mal-ware , bot control traffic etc .. signature extraction * CRM integration * Network event identification * Digital Forensic analysis * etc...

I was quite unhappy regarding the space & processing power that require to store and process the network traffic data into general database. As well as the efficiency when handling the network flow. For an example in an ISP the data generated from Netflow is a significant amount that varies with the number of flow eg:Million flows per hour. Netflow considered De facto standard for network accounting / billing and now anomaly detection schemes. Therefore the general database structures not an efficient for processing large network traces. Therefore i was checking relevant projects in the academic world but AT&T Industry have come up with the project called Gigascope(Still i couldn't get the source code :) ). But similar product i found interesting CoMo Project. This area of researches are follows the "Network Data Streaming" Database structure. Lets explore it "http://como.sourceforge.net/publications.php" :)

Sample network is implemented with similar devices and the Internet connection simulated via Lan connectivity. Two end hosts provide the transfer and reviver functionality. Network Setup. The test environment designed as explained in the diagram 1.0. Test Setup Figure 1.0 End Host Configuration- Table 1.0 End Point 1 End point 2 Microsoft windows XP Professional SP3 CPU count 2 1.87 GB RAM 3.00GHz 100Mbps Realtek RTL8139/810x Family Fast Ethernet NIC Microsoft windows XP Professional SP3 CPU count 1 1.47GB RAM 3.06GHz 100Mbps Realtek RTL8139 Family PCI Fast Ethernet NIC Router Configuration - Table 1.1 Router 1 Speed 264 MHz 264 MHz Memory 30MB RAM 30MB RAM HD 128 MB 64 MB RouterOS 3.30 RouterOS 3.20 Workload: Hashing algorithm parameter is fixed to SHA and changed the encrypt...

This was rudimentary script needed to edit more, but found interesting . #! perl -slw use strict; use Win32::ChangeNotify; use threads; #monitoring path. my $path = 'c:\folder'; my $notify = Win32::ChangeNotify->new( $path, 0, 'FILE_NAME' ); my %last; @last{ glob $path . '\*' } = (); my $count = 0; while( 1 ) { next unless $notify->wait( 10_000 ); # Check every 10 seconds $notify->reset; print $/, 'Something changed'; my @files = glob $path . '\*'; if( @files> scalar keys %last ) { my %temp; @temp{ @files } = (); delete @temp{ keys %last }; #print for keys %temp; my $k; my $v; while ( ($k,$v) = each %temp ) { my $thr1 = threads->create(\&load, $k,$count); } } else { print "A non-deletion or creation change occured"; } undef %last; @last{ @files } = (); $count = $count +1; } sub load { my...

I don't know how many people are still using telnet but suppose if you are using this will help. Script uses export command to take the backup and save it. I thought better to share it. I'll check the ssh variation and upload when possible. #Date - 31st April 2010 #Gobinath #To retrive the backup from the mikrotik router. #!/usr/bin/perl use Net::Telnet; use POSIX ; #Input # routerlist contains ipaddress and name that save the backupfile seprated by tab # used for comments. #Backup Path. $path = "/mikrotik/backups"; #router List $filename = "routerlist.txt"; #Mikrotik Username Password +ct used to remove the color codes @username_list = ('admin+ct','admin'); @password_list = ('######','######'); #Read the router List open (FILE,$filename) ; #Time Stamp. my $datestamp = strftime("%Y%m%d",localtime); chomp($datestamp); sub retriver { undef @output ; my($username, $pass,$hostname) = @_; $telnet = new Net::Telnet (...

Search on gigablast provide a unique facility to search based on IP/IP Range. eg: ip:69.63.181 (http://gigablast.com/search?k4a=576656&s=10&q=ip:69.63.181) List all the websites hosted on the facebook ip range. quite interesting to know. :)