The Hindu Business Line : What's wrong with your network?

WHAT can go wrong will go wrong. Nowhere is this more felt than in the realm of network management. "Servers can crash, WAN links can become saturated, and for unknown reasons, an application's performance can come to a crawl, pitting network engineers against application developers in a complicated blame game, usually without facts," writes Kevin Burns in the intro to his book TCP/IP Analysis and Troubleshooting Toolkit. TCP/IP is the backbone of the Internet and corporate networks, states the back-cover and the book has solutions "that solve real-world networking problems". A sampler:

If one protocol, more than any others, is responsible for the growth of the Internet, it would have to be HTTP. Also known as the Hypertext Transport Protocol, HTTP, like FTP, is a protocol used for the transport of information. But HTTP does much more than just transport information. It has built-in mechanisms for handling different data formats, provides connection persistence, and includes indicators for caching.

Fault management systems are the staple of any corporate network management centre. They usually consist of a large centrally-located computer or computers that actively poll devices on the network to confirm that the devices are still functioning. A standard database called a `management information base' (MIB) allows a management station to query network devices and obtain statistics, such as uptime, utilisation, or error information, from this database. A management station using a protocol called SNMP (Simple Network Management Protocol) can retrieve virtually any piece of information that you can configure the device to store in the MIB.

IP is what is called an unreliable connectionless protocol. Although it is responsible for getting our data from one place to another, it does not guarantee that it will make it.

When TCP sends a segment of data to a host, it will wait for a set period of time called the retransmission time-out (RTO). If it does not receive an acknowledgement from the host during that time, it will retransmit the data segment. Most implementations determine the RTO by something called the smooth round-trip time (SRTT).

Web sites have become very complex. Take, for example, the National Football League's Web site (www.nfl.com) , which contains thousands of statistics, up-to-the-minute score updates, and even live streaming audio of games in action. Today, complex Web sites do not exist as single servers but as multiple multi-tiered systems comprising back-end databases, application servers, and application-aware load-balancing switches.

Put the book on your network before linking the wires.

CLDS, not SDLC

FIRST, the computer guys spoke of bits, then bytes and characters. Thereafter, came fields, records and files. The arrival of databases was a big thing. "Early renditions of databases centred around a single database serving every purpose known to the information processing community - from transaction to batch processing to analytical processing," writes W.H. Inmon in the latest edition of his book Building the Data Warehouse, which caters to the enlightened notion of databases - that separates the operational needs from analytical ones that serve informational requirements and decision support systems (DSS). The book focuses on the `information warehouse' that lies at the heart of DSS processing. A peek into the warehouse:

One confusing aspect of data warehousing is that it is an architecture, not a technology. This frustrates the technician and the venture capitalist alike because these people want to buy something in a nice clean box. But data warehousing simply does not lend itself to being "boxed up".

The development of data warehouse operates under a very different life cycle, sometimes called the CLDS (the reverse of the SDLC). The classical SDLC is driven by requirements; thereafter are the stages of design and development. The CLDS is almost exactly the reverse: it starts with data. Once the data is in hand, it is integrated and then tested to see what bias there is to the data, if any. Programs are then written against the data. The results of the programs are analysed, and finally the requirements of the system are understood. The CLDS is usually called a `spiral' development methodology.

The single most important aspect of design of a data warehouse is the issue of granularity. Indeed, the issue of granularity permeates the entire architecture that surrounds the data warehouse environment. Granularity refers to the level of detail or summarisation of the units of data in the data warehouse. The more detail there is, the lower the level of granularity, and vice versa. For example, a simple transaction would be at a low level of granularity. A summary of all transactions for the month would be at a high level of granularity.

The first impulse of many Web designers is to store Web data in the Web environment itself. But very quickly the Web becomes swamped, and once that happens, nothing works properly. Data becomes entangled in everything - in access queries, loads, indexes, monitors and elsewhere.

Coming to the aid of the Web is the data warehouse itself, as well as the bulk storage overflow component of the data warehouse. Typically, a Web environment holds a day's worth of data, while the data warehouse might hold a year's worth. And the overflow storage component holds as much as a decade's worth of data.

End users operate in a mode that can be called the `discovery mode'. They don't know what their requirements are until they see what the possibilities are. Initially populating large amounts of data into the data warehouse is dangerous - it is a sure thing that the data will change once populated.

A book that you can populate your rack with.

Block by block

ARE you an engineer or designer out there looking for software to do your work? Look at one of the world's "fastest growing solid modelling softwares", says Sham Tickoo of Purdue University in his book SolidWorks Release 2003. The software is a `parametric feature-based solid modelling tool". It unites the 3D parametric features with 2D tools, and also addresses "every design-through-manufacturing process". A few dimensions from the book:

In the `part' mode, you are provided with the standard hole library known as `hole wizard'. You can create simple holes, tapped holes, counterbore, countersink, and so on. The holes can be of any standard such as ISO, ANSI, JIS, and so on. You can also create complicated surfaces using the surface modelling. Annotations such as weld symbols, geometric tolerance, datum references, and surface finish symbols can be added to the model.

Feature is defined as the smallest building block that can be modified individually. A model is a combination of a number of individual features and each feature is related to other features directly or indirectly. These features understand their fit and function properly, and therefore, can be modified any time during the design process.

Collision detection is used to detect the interference and collision between the parts of an assembly when the assembly is in motion. While creating the assembly in SolidWorks, you can direct the collision between different parts of the assembly by moving and rotating the components of the assembly.

An over-defined sketch is one in which some of the dimensions, relations, or both, are conflicting, or the dimension or relations in the sketch have exceeded the required number. In the dangling sketch, the dimensions or relations lose their reference because of the deletion of an entity from which it was reference.

Reference geometry features are those that consist of no mass and no volume. These are available only to assist you in the creation of models. They act as a reference for drawing the sketches for features, defining the sketch plane, assembling the components and so on. Reference geometry is widely used in creating complex models.

A book that performs a dual role - teaching you the software and also a lot of engineering stuff.

Books courtesy: Wiley Dreamtech www.wileydreamtech.com

Please e-mail us on the latest IT books you have read at Books2Byte@hotmail.com

Article E-Mail :: Comment :: Syndication

Stories in this Section
The writing's on the wall

Undaunted by size

Breathing easier

It's how you are seen that counts

Shutdown hitch

Plumping for the Penguin

Shabash Subash

Big push in sight

A time to talk

Quiz

What's wrong with your network?

Cartoon