Conventional Data Center networks have 3-4 tires like access layer, aggregation layer, core layer, services layer, etc. Each layer duplicates many of the IP/Ethernet packet analysis and forwarding functions. This adds end-to-end latency.

To reduce costs and make these networks scalable, over-subscription (higher downlink capacity and lower uplink capacity) is typically used for all these tires. Over-subscription might result in packet losses during network congestion and is not suitable for storage traffic.

STP (Spanning Tree Protocol) is used to block multiple paths (loops) in the network. STP prevents the complete usage of available network bandwidth and the network may have to be rewired to handle changes in application workloads.

Link Aggregation is used to create larger pipes, but it cannot be dynamically modified by the network. Manual intervention is required to create/re-configure link-aggregation to suit the changing application requirements.

Firewalls and load-balancers for each application are dedicated to a group of application servers.

Network management is centered in the switch operating system.

It is difficult to handle Virtual Machine (VM) migration and multi-tenancy (for cloud networks) using conventional data center networks.

Multiple networks maybe used for connectivity (IP/Ethernet), Storage (FC – Fiber Channel), HPC – High Performance Computing (Infiniband), etc.

Modern Data Center networks (Wish-list)

The modern data center network should be able to reduce the 3-4 tires of the conventional DC network, into a simple 2-tire network (access layer and core layer) in order to easily facilitate the increasing traffic between the servers.

Larger layer-2 domains that are easy to scale, facilitate automatic VM Migration and have lower latencies, could be created.

The modern data center architecture should enable people to partition computing, storage and network resources rapidly and dynamically, in order to accommodate for flexible multi-tenant cloud based services, like a private cloud each for different departments within an organization.

The network intelligence could either be placed in the network core, or at the network edge, or within every server (where the application resides). The third option more readily supports the creation of a dynamic logical network that provides for easier application mobility.

Self aggregating logical ISL (Inter Switch Links) could replace manually configured Link Aggregation Groups (LAG’s).

A common management architecture for servers, storage and networking could replace the current requirement for individual management platforms for each of them.

VM’s (Virtual Machines) could be moved from highly utilized servers to under-utilized servers dynamically, along with their network state/storage attributes that move along with the VM’s. Virtualization could also be automated using a common set of standard tools.

Modern data center networks should be able to be scale dynamically (even with live traffic) using a set of modular hardware and modular software components, while still maintaining low-cost per port.

A network is generally over-subscribed in order to operate/expand in a cost-effective way. But there are certain types of traffic in the modern data centers (like storage – FCOE traffic) that cannot tolerate packet losses due to over-subscribed networks. Hence, it is important to introduce some kind of throttling mechanism in modern data center networks to pause traffic from the source, in case of network congestion.

Network cables can be connected once and dynamically reconfigured using SDN – Software Defined Networking.

This post represents the highlights of the document, ‘ODIN Volume 1: Transforming the Data Center Network’, by IBM’s ODIN (Open Data Center Inter-operable Network) initiative. You can read the complete discussion on this topic by downloading the document from the above link. This is a part of the ODIN series of posts.

excITingIP.com

You could stay up to date on the various computer networking/IT technologies by subscribing to this blog with your email address in the sidebar box that says,’Get email updates when new articles are published’

When companies are building large cloud-scale data centers, it is important to consider open-standards and multi-vendor compatibility while designing the network architecture. Even though IBM offers its own line of networking solutions for the data center, I think the ODIN is a good initiative and a number of other networking vendors also seem to think so.

These are some of the objectives of designing a network according to ODIN principles:

• Scaling of networks to thousands of physical ports without over-subscription
• Achieving virtual machine mobility using layer-2 domains
• Achieving an ultra-low latency over WAN
• Maximizing high-availability, energy efficiency
• Automated management of network systems
• Reducing capital and operating expenses

ODIN is not a standard in itself, but it establishes a platform and helps customers to adopt existing open standards in large-scale/data center networks.

The following technical brief documents have been released by IBM. You can download them from here.

1. Transforming the data center network
2. ECMP Layer-3 networks
3. Software Defined Networking and Openflow
4. Lossless Ethernet
5. WAN and ultra-low latency applications

I’ll give a short overview of each of them in this blog over the next few days and link those pages above. If you want to read the complete brief, you can download the documents from here and read them.

excITingIP.com

You could stay up to date on the various computer networking/enterprise IT technologies by subscribing to this blog with your email address in the sidebar box that says, ‘Get email updates when new articles are published’

We are all familiar with Laser and Inkjet printers that can print text or images on paper. But there is something called as a 3D printer that can ‘print’ real objects and materials if a 3D input source is provided to it! The input can be a 3D scanner or 3D modeling software file from Google Sketchup or Auto-CAD. How cool is that?

The above embedded Youtube video shows how a 3D printer prints out a ‘duplicate copy’ of a tool after scanning it using a 3D scanner. Don’t miss

3D printers basically use ‘additive’ manufacturing technology to create a three-dimensional object layer by layer (one layer at a time). This is different from the ‘subtractive’ manufacturing process (normally used) where the unwanted pieces are removed from a block of material to create a tool of a certain specifications. They can even create tools with moving parts with just a single ‘print’.

Most of them use powder based printing technique or polymer based printing technique to create new objects. The duplicate object might be made of a totally different material or the same material, depending on the type of printer used. It is possible to print customized items in plastic, ceramic, glass, stainless steel and even precious metals.

It seems this technology was available for over three decades and was used to make prototypes of tools and materials required in manufacturing companies. But it was quite expensive. However, 3D printing has gained a lot of popularity in recent times and people have started using it in their offices/homes due to their falling prices.

The real value of 3D printers is achieved when a user wants to manufacture some material using their own custom design or a machinery unit wants to manufacture something in minor quantities (like a prototype). Otherwise, 3D printing can be very expensive.

Some companies provide 3D printing as an online service. That way, you can design a three-dimensional object, upload your image file and order a ‘print’. It is shipped to your house in a week or so, when ready. You can also choose from the designs already created by various members.

There are some open-source based 3D printer models available at affordable prices. These printers might have limited capabilities but they are a good way to start with 3D printing.

You can read more technical/general information on 3D printers from this Wikipedia article.

Research is very much on to recreate or ‘print’ food items, medicines, motherboards, processors, bones, chemical compounds and every other possible material using 3D printers. If it succeeds (many people believe so), it has the potential to change the way we live! Let us wait for a few more years to find out if 3D printing technology becomes the next big technological wave that affects all our lives, like the Internet.

excITingIP.com

The following certifications are planned for individuals:

• Certified developer
• Certified migration professional (MS Office to Libre Office)
• Certified professional trainer
• Certified L1 support professional
• Certified L2 support professional

You can get more details about the certification process in general and the above individual certifications from this page.

System Integrators, solution providers and resellers can now start looking at providing services/support based on LibreOffice. Though the software is free for end-users, System Integrators can still provide services like installation, migration, training and support. System Integrators can even display a ‘Libre Office certified’ badge if they employ or contract certain number of Libre Office certified professionals/ LibreOffice contributors, based on the total number of employees. Many vendors are already providing services for LibreOffice.

The certification process is to be overseen by a committee appointed by The Document Foundation and the training will be initially managed directly by the foundation. There is a modest fee applicable for these certifications which need to be renewed every 24 months. Further specifics on the certifications/process is to be discussed during the first meeting of the certification committee at LinuxTag (Berlin, May 23 – 26).

For end-users (enterprise/corporate customers), this will provide a good opportunity to migrate from proprietary office suites and implement open-source technologies duly supported by partners/System Integrators. Lack of support has been a factor preventing bigger companies making the move, yet. But with these developments, companies could rethink their enterprise office productivity suite strategy.

excITingIP.com

• An administrator performing remote maintenance activities over LAN need not switch On or switch Off individual machines physically. They can just remotely switch it On, complete the maintenance activities and switch it Off, without leaving their desk.
• Some servers need not be always On. They can normally be kept in the switched Off state and switched On from anywhere, when required. This saves a lot of power, reduces the heat & wear and tear due to continuous operation.
• Sometimes, we leave a computer On just to access files from them over the Internet (from somewhere else), after a few hours. WOL enables us to shut-down the computer or keep it in hibernation until it is switched On from a remote location (only when required). This saves power, cooling and it is better for computers because they are generally not optimized for 24×7 operation.

A system being switched Off refers to the computer/server being shut-down, but the main AC power cord should still be connected to an electrical source and it should be switched On. This is required because, though the computer itself doesn’t draw any power when it is shut-down/in hibernation, the NIC (Network Interface Card) needs power to be able to continuously receive all the packets coming its way and recognize the special ‘magic packet’ that instructs it to start a system.

For WOL to work, it should be integrated into Network Interface Cards(NIC)/On-board NIC in the motherboard. Further, WOL needs to be enabled in the BIOS. Once the ‘magic packet’ containing the MAC address of the system to be switched On is received by the NIC (either from the LAN or from the WAN/Internet), it starts the procedure to switch On the system. Switching On refers to loading the BIOS, operating system and applications.

To use WOL, one might require administrative access to the system. WOL does not replace any user-names/passwords required to access the system and once switched On, it is similar to remote desktop that allows remote access to a system. A tutorial on how to enable Wake On LAN can be found here. Another tutorial for enabling WOL over the Internet (WAN) can be found here. It is even possible to Wake On Wireless LAN (WoWLAN) that uses the 802.11 Wi-Fi network to switch On or switch Off a system remotely.

There are a few limitations to the WOL protocol. For example, since WOL is a receive only protocol, there is no mechanism to send a confirmation packet, back to the initiator. Also, WOL is a broadcast protocol. So, routers should be configured to forward broadcasts.

excITingIP.com

You could stay up to date on the various computer networking/enterprise IT technologies by subscribing to this blog with your email address in the sidebar box that says, ‘Get email updates when new articles are published’

]]> http://www.excitingip.com/2975/you-can-use-wol-wake-on-lan-to-remotely-switch-on-or-off-systems/feed/ 0 http://www.excitingip.com/2969/internet-on-mobile-is-fun-its-more-work-actually/ http://www.excitingip.com/2969/internet-on-mobile-is-fun-its-more-work-actually/#comments Tue, 01 May 2012 03:02:04 +0000 admin http://www.excitingip.com/?p=2969 This blog, unlike my other blog, has been consistently technical and serious. So, I thought why not break the monotony and write something lighter?

Before about 3-4 years, I got my first smart phone – HTC p3400i. You may not believe, but I am still using it! Anyways, going back to those days, 3G was yet to be released and we had only EDGE. I thought something is better than nothing, and bought it especially for Internet browsing (due to its large touch screen).

I used to follow a lot of blogs and I wanted to be able to read and comment on them using my mobile phone. In my line of work (back then), I had to wait endlessly at certain places and I also had to travel quite a bit. So, I thought having a touch screen phone with Internet access might make things more fun.

It was, but each action has an equal and opposite reaction, right?

My first mistake was flaunting my new smart phone in the office. Obviously, my boss got to know about my new phone, and its capabilities. From that time, he used to casually call me, where ever I was and ask me to send mails to anyone at anytime! Before that, I used to call him and asked him to send mails on my behalf!

People are social animals. They like to share good things. So, my boss tells all our major customers that I have a touch screen phone with Internet access. So, once I get a call from a customer,

………………..

“Can you please confirm this by mail as soon as possible?”

“Sorry sir, I am outside now. I will go back to office in the evening. I will send a mail by then”

“Ha ha. I know you have a new touch screen phone with Internet access. Just send a mail from your mobile”

“!!!!!”

Then, the news spreads even to my Head Office. Someone from head office calls,

………………..

“Why don’t you send a follow up mail on this matter immediately, marking a copy to us?”

“I am outside right now”

“I know you have a new mobile with touch screen and Internet”

“!!!!!”

At that time, I wouldn’t have been surprised if the news of my new touch screen mobile with Internet ‘capabilities’ was featured on the front page of the local news paper. It was so popular!

Till then, I just had intrusion in the form of calls to my mobile. It was enough if I promised people that I would complete my work once I reached office. But now, I had a mobile office in my pocket! That means, work anywhere and everywhere!!!

Of course, there is an equal and opposite reaction to work, as well. So, I got to read all my favorite blogs and comment on them in-spite of these unwarranted interferences!

How do you use your mobile? For FUN or for WORK?

excITingIP.com

This post is my entry for the indiBlogger.in contest ‘Internet is fun’, sponsored by Vodafone. Readers can have a look at the Vodaphone fun page – http://www.vodafone.in/fun

Should IT managers be concerned about Computer Forensics? Yes, they should. Tracing and preventing an attack from harming an organizational network is one of their objectives. Though UTM/IPS systems can secure the network to a certain extent, in some situations it is required to trace out and identify the real identity of the attacker, and bring them to justice. That way, repeated attacks by a particular attacker can be stopped.

IT managers/Network administrators might be surprised at the amount of information that can be retrieved from internal and external systems, post attack. It is always better to be plan and be ready so that when security breaches actually happen, it might be easier to trace the attacker. Let us discuss some methods of tracing down online attackers.

There are tools (1, 2) that can help trace an email or an IP address back to its sender and provide a good amount of publicly available information about the sender. This is important because there are so many phishing spam mails that are sent to dupe users with malicious links that try to extract financial/personal information. An email header (present in every email) contains information about the origin of the message, all the systems it has passed through (with date and time) and other such valuable information, which can be extracted and analyzed by certain programs.

If an insider attack is suspected or if the computer used by an external suspect can be accessed, it is possible to trace out all their Internet activities like websites visited, files downloaded, etc from a file called index.dat, which is stored securely in most of the common browsers (it is usually difficult to delete). There are index.dat viewers/analyzers that help in extracting and documenting the details of the web-activities of a suspect/attacker.

Even if an attacker uses multiple intermediary systems to carry out their attack, it is still possible to trace back the attack to the initiator or at least the Network/ISP of the initiator. This can be done because routers present across the Internet have their own markings, record routes taken by attack packets and all that information can be traced  by using appropriate tools. If the identity of the intruder cannot be traced directly, some coordination with the ISP should reveal the identity of the attacker.

Some companies set-up honeypot systems that deliberately run software with known vulnerabilities. These systems are kept to entice attackers to hack into them so that the attack pattern can be studied and perhaps the source of the attack can be determined. These systems are isolated from the main network, so that other systems are not affected by such attacks.

Logs are maintained by almost every IT equipment deployed in a company. Servers, firewalls, UTM, IPS and many more devices/applications store retrievable logs that can be useful to study a security breach, when it occurs. SIEM (Security Information and Event Management) systems enable centralized collection and analysis of logs from various security devices across the network and this could be useful for analyzing and tracing an attack.

There are various methodologies adopted by various organizations, government agencies and even private agencies to trace online attacks and prove the misconduct in a court of law. IT managers/Network administrators should be aware that such things can be done and have appropriate systems in place in order to conduct/help the investigation process after a security breach.

excITingIP.com

You can keep yourself updated with the various computer networking/IT technologies by subscribing to this blog with your email  address in the sidebar box that says, ‘Get email updates when new articles are published’

Have you ever thought of creating an exclusive 5 Ghz Wi-Fi network? It’s no more an impractical idea, especially since the 2.4 Ghz spectrum is getting over crowded and 5 Ghz is the better option for channel bonding, which is key to higher 802.11 n performance. If you build a 5 Ghz network, your neighbors will love you

But practically, building a Wi-Fi network exclusively on the 5 Ghz band may not be possible because in larger wireless networks, there are so many wireless clients that may not have 5 Ghz capable wireless adapters. High end laptops and ipad may come with it by default, but most other devices like Wi-Fi enabled cell phones, Wi-Fi printers, etc may not.

Another hurdle to building a 5 Ghz network is its cost. 5 Ghz capable Access Points and Wi-Fi adapters are more expensive than their 2.4 Ghz counterparts. But as the popularity of 802.11n networks increase, we can expect more and more 5 Ghz capable clients and access points to be dual band enabled and cost lesser.

2.4 Ghz, being a lower frequency band offers a higher range (distance coverage) than 5 Ghz. But this can also be an issue in large networks/dense networks as multiple access points within a certain range can cause more interference.

But in-spite of these limitations, 5 Ghz Wi-Fi networks are popular because of the following advantages.

Wi-Fi: Advantages of 5 Ghz Network Vs 2.5 Ghz Network

Depending on the country and use of channel bonding, 2.4 Ghz spectrum has 1 – 4 non over lapping channels and 5 Ghz spectrum has 8 – 24 non over lapping channels. Data in Wi-Fi is transmitted on non-overlapping channels in order to avoid interference and the resulting packet collisions/packet loss. With so many non-over lapping channels, 5 Ghz band is the best choice for larger networks as it can avoid interference from other AP’s in the vicinity, much better than 2.4 Ghz band.

5 Ghz spectrum is capable of transmitting at a higher data rate (higher bandwidth) due to a technique called ‘Channel bonding’ used in 802.11n. In this technique, two 20 Mhz channels are combined to form one 40 Mhz channel and hence the bandwidth capacity is almost doubled. But, use of channel bonding cuts the number of available channels into half and hence it can be more practically implemented in the 5 Ghz band.

The 2.4 Ghz spectrum has a lot of interference not only from neighboring access points in the same floor (and the ones on the top floor/below floor), but also from other devices/wireless standards operating in the same frequency. For example, Bluetooth, cordless phones, wireless speakers, microwave ovens, baby monitors, etc could interfere with Wi-Fi signals. This results in reduced efficiency as packets lost due to collision need to be retransmitted/recreated. There are no such problems with the cleaner 5 Ghz band which is mostly free from interference.

Certain vendors claim that 5 Ghz radio is better as it can reflect more and scatters better. So, these waves can travel around obstructions (through reflections from neighboring surfaces). But there are other vendors who claim just the opposite of this, that 2.4 Ghz spectrum can travel through obstructions better. I am not sure which one is true and the reader can comment if they have a better idea on this topic.

5 Ghz seems is a better choice with larger wireless networks because of the additional non-overlapping channels present in it. It is possible for network managers to carefully plan the frequency of operation of adjacent access points (and the ones on the top floor, bottom floor) so that they don’t interfere with each other.

Dual band Wi-Fi implementations:

Practically, the best option available for network administrators is to go for a dual-band Wi-Fi implementation. This is because there will always be some legacy clients that do not support 5 Ghz and guests visiting the campus may not have 5 Ghz adapters on their laptops. Dual band implementations are generally planned for 2.4 Ghz with an overlay of 5 Ghz network.

In order to increase the efficiency of dual band networks,

Devices that can operate in 5 Ghz can be forcefully made to operate in 5 Ghz (both AP’s and clients can be set to operate only in 5 Ghz) (or) the controller/access points can push them over to 5 Ghz in order to free the available bandwidth in the 2.4 Ghz band for other devices. The second technique is called band steering and certain controllers can automatically do it when a higher number of devices start connecting in 2.4 Ghz.

Some access points allow software selection of frequency bands in individual radios and this could be very useful as the frequency of operation of each radio could be decided based on the load/capability of devices, at any point of time.

One more technique used by administrators is to lower the transmit power of 2.4 Ghz radios, so that dual band enabled wireless clients will automatically connect to the high power 5 Ghz radios.

802.11 ac is the next (new) wireless standard that is coming up and it primarily operates in the 5 Ghz spectrum. This should encourage more companies to migrate to 5 Ghz gradually. It is to be seen if 2.4 Ghz is supported with it, as well.

excITingIP.com

Have a look at this video to get an idea about how the Kensington Lock works:

When going through the specifications for laptops and projectors, I have noticed a ‘Kensington slot’ being present in some of them. I have always wondered what this Kensington slot/Kensington lock is, until today.

Kensington is a company that has been manufacturing physical locks for laptop (and other electronic equipment) security for quite a long time and hence many vendors have Kensington slots integrated with their electronic products. You can see Kensington slots in Laptops, Desktops, Projectors, Printers, Servers, Flat panel monitors, etc.

Kensington locks are made of a single click locking mechanism and high carbon steel cable attached to it. It can lock your electronic equipment to a rod, table leg, window frame, etc by pushing the lock over an attachment placed on the Kensington slot (single click locking mechanism) or it can be locked manually by using physical keys.

One can open/release the lock using a physical key or a number combination (available in certain locks). The company even offers a web-registration portal to register your keys, so that in case you lose them and the laptop is locked, they can send duplicate keys to you!

Kensington slot is a standard hole that is already integrated in many electronic equipments. A Kensington lock can fit into this hole and with another attachment, it can just lock-in when the user pushes the lock into the hole (see the above embedded video to get a good idea on how to lock a laptop physically).

The Kensington physical security lock can be used in many situations like a college dorm, offices, client offices, transport facilities (trains/buses), coffee shops, conference rooms and anywhere else where one might leave the laptop/other electronic equipments unattended for a considerable amount of time. It can prevent opportunistic thieves from running away with your laptop/projector.

excITingIP.com

The above video is a demo video for something else, but they do show glimpses of how two monitors can be connected to a video conferencing system, where one displays video and another displays data/presentation. Sorry, I didn’t get a video specific to H.239!

What is H.239 in Video Conferencing?

H.239 is a standard to share data/presentations/second video channel along with the primary video channel. That means, while you can see and talk to people from remote locations using video conferencing technology, you can also share your computer monitor output or a document camera output to the participants on the remote end. The person who is talking can show a presentation/ graphic file from their computer, to remote conference participants. This enables people to understand the context of discussion better.

H.239 basically enables a video conference system to send two streams of media together. So, either video is combined with computer monitor output (data/presentation) or an auxiliary video source and sent to the other end, which can divide them into two separate streams and display them separately. People on the remote remote end can view the computer monitor/document output connected on the other side, LIVE. If a cursor is moved from one location on the computer screen to another (for example), the people on the remote end will be able to see it in real-time.

Each VC system supports a certain maximum resolution for data transfer (Like HD, XGA, SVGA, etc). Since two streams of media are transmitted with H.239, some additional bandwidth is occupied but the maximum bandwidth limit for the data stream can be set in many video conference systems.

Which media sources can be connected to a video conference system & shared with the remote end?

Many video conference systems support a PC/laptop to be connected to it (over a VGA/DVI cable).  Some of them can even connect to a document camera/DVD player/second (auxiliary) camera, etc and send their output to other end. These additional multimedia streams are sent along with the main video output of the video conference system.

How do people at the remote end view the data/presentations/auxiliary video?

Depending on the make and model of the video conferencing system, there are many ways to view the second media stream. The best way is to have two different monitors or one monitor + one projector connected to the VC system and display the main video (people) on one of them and display the data/presentations on the other. This enables people to easily see both outputs clearly.

But some video conferencing systems do not support two monitors to be connected to them. In such cases, people can use a single monitor but they need to toggle between the video and data using their remote control or they can view them side by side on the same screen (screen is divided into two parts).

Some VC systems support picture in picture feature which lets the users see the presentation on the main screen and the people (main video) on a small box on one corner of the same screen. There are multiple customizations available for these options.

How does one enable H.239 in their VC system?

While most of the video conferencing systems meant for enterprises support H.239, a few models may not. So, its better to check before-hand. Many VC systems support this feature natively but some of them require additional software licenses or additional hardware units to enable H.239 (at additional cost).

Once this feature is enabled, it is just a question of connecting a media source and sharing its output by pressing a key on the remote control. It is better if both the video conferencing systems that want to use H.239 are manufactured by the same vendor. But it is possible to use it, even otherwise.

Compatibility:

Since H.239 is a standard, the data sharing capability of the VC system is supposed to work even if there are different brands of video conferencing equipments at either end. While this is possible, there are factors like make, model, firmware version, license and other factors that can hinder VC models manufactured by different vendors to share data/content using H.239.

H.239 is supposed to be compatible with Video Conferencing systems connected over both IP and ISDN networks. In case of Multi-Conference system (video conference between 3 or more locations simultaneously), a user present at one location can connect their laptop and send the data/presentation, which is received by all the participants. Of course, there are some limitations in these cases.

excITingIP.com

You could stay up to date on the various computer networking/ enterprise IT technologies by subscribing to this blog with your email address in the sidebar box that says, ‘Get email updates when new articles are published’