(Brien Posey, TechRepublic) Server virtualization is becoming increasingly popular, and it seems that everyone is in a mad dash to virtualize their datacenter.  While there’s no disputing the benefits of server virtualization, there are some questions you should address before you begin to virtualize your servers.

Virtualizing your servers offers significant advantages, but effective planning is crucial to your success. Make sure you have satisfactory answers to these key questions before you get underway:

1. Does the virtualization plan include a single point of failure?
2. Are all the applications supported in a virtual environment?
3. Are there any servers that are not good virtualization candidates?
4. How will domain controller placement work?
5. What is the most suitable virtualization platform?
6. What is the contingency plan if a host server dies?
7. How many guest machines can each host accommodate?
8. What software licenses will be required?
9. How will the old server hardware be used?
10. What is the plan for existing server clusters?

For more details follow the "full article" link.

(Andy Moon, TechRepublic) A new survey found that only 8% of businesses (small, medium, and enterprise) have any current plans to utilize cloud storage and only 3% are using it now. These results suggest that, while there is a lot of potential for cloud storage, concerns about privacy, security, and pricing are keeping most companies from moving data out of their data centers, at least as a primary storage option.

Forrester analyst Andrew Reichman writes in the report that “there is long-term potential for storage-as-a-service, but Forrester sees issues with guaranteed service levels, security, chain of custody, shared tenancy, and long-term pricing as significant barriers that still need to be addressed before it takes off in any meaningful way.”

One interesting finding of the survey is that companies are more interested in the cloud for back-up storage rather than general purpose storage, which requires a much smaller leap of faith and is far easier to implement than remote primary storage.

Lately, more and more companies that rely on IT services have begun to understand the true benefits of network subsystem interaction. Some of the well established examples are already accepted as being the de facto way of doing things. To a lot of people, the Apache integration with MySQL or PHP is no longer considered to be a bunch of independent software packages that cooperate. This cooperation is so productive that everything got merged into a single concept: the AMP (i.e. LAMP, WAMP etc.).

Along the same lines, the AXIGEN electronic messaging system can be expanded to provide external applications with the information required to perform advanced tasks in a very productive and cost effective way. You should bear in mind that some of the expansion methods described in this article are generic and can be applied to various situations while others are very specific.

(ZDNet Australia) Server virtualisation is a no-brainer -- it's quick to deploy and easy to justify in terms of cost-savings, but too many companies are deploying the technology without considering the security implications.

Server virtualisation has been the hottest trend in enterprise IT for some time and according to IBRS analyst Kevin McIsaac, it's likely to remain that way for the next two to three years.

IBRS estimates that one in three large Australian organisations has deployed server virtualisation within their datacentre, and nearly every medium to large enterprise has at least looked at a pilot for the technology.

In today’s business environments, we often hear the term ‘high-availability’. Whether a hardware or software solution, we aim for redundancy at all levels, so as to maximize the availability of mission-critical services and operations.


1. Introduction
This white paper discusses a high-availability solution for the AXIGEN Mail Server, using the Heartbeat package with DRBD.

AXIGEN is a proprietary messaging solution while both Heartbeat and DRBD are open source software released under the GNU Public License (GPL). These software packages can be used together to build scalable and highly available integrated cluster messaging applications on the Linux operating system.

Two in five small to medium businesses (SMBs) say virtualisation is the best way to bypass a potential recession.

Of 100 companies questioned by internet service provider Star, 43 per cent cited virtualisation tools as a key way to protect their organisation from the current economic climate. A further 21 per cent of respondents said that a converged approach to new technologies was the best strategy for dealing with the situation.

The availability of broadband was the most important factor for 17 per of respondents, while 10 per cent believed new Web 2.0 technologies would play an important role in battling economic instability.

"One of the biggest challenges facing small businesses today is access to finance, so as the economic slowdown continues, the number one priority is to do more with less, making the most of existing technology investments and using technology to respond quickly to changing business conditions,” said Ben White, chief executive of Star. “Virtualisation helps achieve those business needs."

The ability to 'work smarter' was the most powerful driver behind making new technology purchases, as cited by 43 per cent of those questioned.

Demilitarized zone, used to secure an internal network from external access. You can use Linux firewall to create DMZ easily. There are many different ways to design a network with a DMZ. The basic method is to use a single Linux firewall with 3 Ethernet cards. The following simple example discusses DMZ setup and forwarding public traffic to internal servers.

Consider the following DMZ host with 3 NIC:
[a] eth0 with 192.168.1.1 private IP address - Internal LAN ~ Desktop system
[b] eth1 with 202.54.1.1 public IP address - WAN connected to ISP router
[c] eth2 with 192.168.2.1 private IP address - DMZ connected to Mail / Web / DNS and other private servers

  

Here is a thorough article about emails, put together by Marshall Brain and Tim Crosby. Inside the article:
Introduction to How Email Works; A Simple Email Server; More Complex Servers; The SMTP Server; The POP3 and IMAP Servers; IMAP Problems and Attachments; Free and Paid Email Services; Email Etiquette, and others.

Introduction to How Email Works:
Every day, the citizens of the Internet send each other billions of email messages. If you're online a lot, you yourself may send a dozen or more emails each day without even thinking about it. Obviously, email has become an extremely popular communication tool. Have you ever wondered how email gets from your computer to a friend halfway around the world?

Running your own email server can grant the owner a significant amount of freedom and experience.  It allows you to set up and maintain multiple email accounts and domains with ease and allows for keeping unwanted mail out of your inboxes.

Also, you can create email distribution lists to allow for simple delivery of messages to a predefined list of addresses, or have an unlimited (up to your hard drive size) amount of space to store mail. You also have full control over what types of spam control, virus protection and relay black lists are protecting you email server. You can also determine what types of protocols users can access their email with.

In this article I will assume you are running an email server on a direct Internet broadband connection with a single WAN IP address. I will assume that you have a server set up and understand the basis behind port forwarding and basic networking skills.

DNS Considerations
In order for other email servers on the Internet to know where to deliver mail for your domain, you must make certain entries in your DNS servers. The most important of these is the MX record. The MX records tell the names or IP addresses of the servers that are allowed to handle email for you domain. You can have multiple records of this type as each one must be assigned a weight or authority that determines the order in which each is contacted to receive mail.

Ages ago (in Internet time), when mainframe dinosaurs roamed the Earth, a new approach to computer networking called "client/server" emerged. Client/server proved to be a more cost-effective way to build many types of networks, particularly PC-based LANs running end-user database applications.

Client/server is just one approach to distributed computing. The client/server model has been popular for a long time, but recently peer-to-peer networking has re-emerged as a viable alternative. Other approaches like clustering also have benefits in specific situations.

Client/server is a computational architecture that involves client processes requesting service from server processes. Some of the most popular applications on the Internet follow the client/server design:
    * Email clients
    * FTP (File transfer) clients
    * Web browsers ....

In this article, we will describe some of the most important things that need to be taken into account when designing a large scale messaging setup.

We are referring to large scale setups, not cluster setups; the main difference being that cluster setups involve a high-availability situation, which is not the case with large scale setups that achieve high availability or redundancy through other means than the mail server software or the way the setup is organized and performed.

In today’s fast paced world, the importance of having an e-mail service needs no further explanation. We all use it on a daily basis, relying on it to keep in touch with the surrounding world. However, when the throughput of information becomes so great that a simple system cannot keep up with it, new methods must be deployed to cope with the growing requirements of networks and their users. Clusters fill up the gap between the need for faster, more powerful systems, and the safety, reliability and easy management of the already existing networks.

Here is an article I found on SearchSecurity.com (TechTarget), about how to prevent spam bots from hijacking an enterprise network. I think you will find it interesting and useful:

"Despite Bill Gates' assertion in 2004 that the problem of spam would be solved by now, it's still with us. In fact, it's on the increase. According to recent figures from Symantec, 61% of emails are spam, and almost 90% of emails emanating from some countries are spam.

This article represents a small guide to understanding the DMZ (Demilitarized Zone) concept and the ways to implement it on a server. It also includes useful tips on how to increase the security of such setups, with a special focus on the Linux/*NIX-type systems:

Common setups used for small and medium networks include a firewall that processes all the requests from the internal network (LAN) to the Internet and from the Internet to the LAN. This firewall is the only protection the internal network has in these setups and it handles any NAT (Network Address Translation), forwarding and filtering requests as necessary. In most cases, the firewall also runs public services accessible from the Internet, such as web services and e-mail services. Within such setups, the DMZ is thus installed on and limited, we may say, to the server.

(Ellen Messmer, Network World) Is moving to virtualization  and cloud computing making network security easier or harder? When some 2,100 top IT and security managers in 27 countries were asked, the response revealed a profound lack of consensus, showing how divided attitudes are within the enterprise.

The "2010 State of Enterprise Security Survey - Global Data" report shows that about one-third believe virtualization and cloud computing make security "harder," while one-third said it was "more or less the same," and the remainder said it was "easier." [...]

The survey showed that the median annual budget for enterprise security in 2010 is $600,000, an 11% increase over 2009, with yet another 11% increase anticipated in 2011.[...]In fact, 40% of the respondents indicated their organizations were currently using applications in the cloud in some way -- yet 40% said it would be more difficult to prevent or react to data loss under their firm's cloud-computing strategy.[...]

(Bill Claybrook, TechTarget) Let's explore the pros and cons of running multiple vendors' platforms in your virtualization environment and review the virtual appliance and cloud computing markets.

Virtualization environments: To mix or not to mix?
If you have some combination of Windows, Red Hat Enterprise Linux, SUSE Linux Enterprise Server, Solaris and Oracle in your data center and you follow our virtualization selection recommendations, your environment will support multiple virtualization software platforms. Each platform has its own set of virtualization management tools, which creates licensing and maintenance complexity.

The survey results indicate that 55.8% of respondents want to standardize on a single x86-based hypervisor if possible. This data aligns closely with the results of a winter 2009 study by The Info Pro, in which 54% of respondents said they do not want to use multiple x86 virtualization approaches. But 54% of respondents said they would use multiple approaches to get the required performance and support.

Realistically, you need at least two or three for high-performance production applications and timely support for guest operating systems when issues arise.

But if you want to standardize on one virtualization platform, VMware stands out as the only solution that is suitable, according to The Info Pro study. IT managers believe in VMware because it dominates the virtualization platform market.

Market-by-market breakdown
There are multiple virtualization software markets: server virtualization, cloud computing, virtual appliances and so on. Microsoft and VMware have the markets for server virtualization and internal cloud computing, but where are the markets for Citrix Systems, Oracle, Red Hat and other open source virtualization software vendors for public clouds and virtual appliances? To decide which virtualization platform vendors to go with in a specific market, you should focus on the size of their installed bases.

(Pam Derringer, News Contributor, TechTarget) Windows job-seekers may have 30% more available openings today in the U.S. than their Linux counterparts. On the flipside, however, the number of Linux listings nationwide is growing while demand for Windows staffers is dropping.

According to the latest figures from IT job site Dice.com, Windows had the most OS job listings on October 1 with 8,865 postings, but the tally was 8% less than the 9,643 Windows opportunities that had been advertised at the start of the year.

Linux ranked third (behind Unix) with 6,168 postings in October. But in contrast to Windows, Linux postings grew by 6% during the year, increasing from 5,839 jobs listed on January 5.

Specific listings for Red Hat Inc., meanwhile, increased nearly 4% from last January, but postings for Novell SUSE Linux Enterprise skills dropped 14% during the same period.

Demand for Windows and Linux platforms, however, was down significantly from a year ago on the cusp of the recession: Windows dropped 43% from last October's total of 15,566 and Linux declined 35% from 9,451 a year ago. Red Hat and SUSE also declined sharply from the previous October, Red Hat by 34% and SUSE by 57%.

(Logan G. Harbaugh, TechTarget) It seems to be assumed that running eight OSes on a single large server with a virtualization suite is cheaper than running eight separate physical servers. Leaving aside for the moment the cost of virtualization software, training admins to use the software and so forth, let's look specifically at the cost of the hardware.

Eight 1U dual-core servers with 4 GB RAM, one or two built-in Ethernet ports and 70 to 150 GB of hard disk space can be found quite inexpensively these days, with even more savings if you don't insist on a first-tier server manufacturer. On the other hand, a 16-core server with 32 GB of RAM, several four-port Ethernet cards and a terabyte or two of disk space can cost considerably more than eight dual-core servers. A quick check of a major brand site produced numbers of $1,339 each for the small servers, or $10,172 total, versus $45,196 for a single big server. If the servers are running mission-critical apps and you want failover capacity, you'll need two big servers for $90,392, rather than nine small ones for $11,511.

If you do need failover, you'll need premium virtualization software in addition to the failover software, which is a considerable expense. With hardware and software factored in, eight virtual servers could be $30,000 to $50,000 more expensive. On the other side of the balance sheet, a virtualized infrastructure can be more flexible and easier to administer, since all eight virtual servers, backups, and failover can be managed from a single console.