Live Chat Software by Kayako
 News Categories
(19)Microsoft Technet (2)StarWind (4)TechRepublic (3)ComuterTips (1)SolarWinds (1)Xangati (1) (27)VMware (5)NVIDIA (9)VDI (1)pfsense vRouter (3)VEEAM (3)Google (2)RemoteFX (1) (1)MailCleaner (1)Udemy (1)AUGI (2)AECbytes Architecture Engineering Constrution (7)VMGuru (2)AUTODESK (1) (1)Atlantis Blog (7)AT.COM (2) (1) (14) (2)hadoop360 (3)bigdatastudio (1) (1) (3)VECITA (1) (1)Palo Alto Networks (4) (2) (1)Nhịp Cầu đầu tư (3)VnEconomy (1)Reuters (1)Tom Tunguz (1) (1)Esri (1) (1)tweet (1)Tesla (1) (6)ITCNews (1) (1) Harvard Business Review (1)Haravan (2) (1) (3) (3)IBM (1) (2) (1) (6) (1) (1) (4) (1) (1) (1) (1) (1) (1) (1) (5) (4) (1) (1) (1) (1) (2) (7) (1) (1) (1) (1) (2) (1) (2) (2) (2) (1) (7) (1) (1) (1) (1) (1) (1) (1)Engenius (1) (1) (1) (1) (1) (3) (6)
RSS Feed
Top 5 Open Source Alternatives to Microsoft Exchange
Posted by Thang Le Toan on 08 February 2017 04:18 AM

Top 5 Open Source Alternatives to Microsoft Exchange

Are you tired of overpaying for the Microsoft Exchange messaging server? The Linux and Free/Open Source world is cram-full of robust, capable alternatives that won't drain your bank account. This tasty assortment ranges from free with community support, to full commercially-supported products. Let's take a look at five open source Microsoft Exchange killers.
Choosing an open source messaging server can save you money and admin time without losing out on features. Take a look at what our favorite alternatives to Microsoft Exchange have to offer.

1. Ubuntu Server

The popular Ubuntu Linux, famous for its polished, easy-to-use desktop version, also has a specialized Ubuntu server edition.This is more than a re-packaged Ubuntu Desktop; it comes with a kernel tuned for server duties, a full complement of server software and no graphical interface. This is a serious, lean mean server operating system.

Ubuntu Server does little handholding, so it's not for novice system administrators. Rather, it is for the experienced sysadmin who wants maximum control. It includes a number of enterprise-ready features such as AppArmor for security, private cloud, public cloud, high availability, a Landscape client for systems management, Mac OS X and Microsoft network integration, and virtualization.

The installer serves up a number of useful package groups such as Virtual machine host, Samba file server, LAMP stack, Printer server, and Mail server. The Mail server group installs Postfix (SMTP), and the excellent Dovecot for POP3 and IMAP4, which is a stout foundation for an on-site email server. Then add whatever additional services you want, such as a Webmail server, calendaring, shared contacts, discussion boards, file shares, and so on. Or you can install a prefab groupware suite on it.

Ubuntu Server is free of cost, and various commercial support and systems management options are available.


2. Citadel Groupware Server

Citadel is free of cost and 100 percent Free software, comprised entirely of GPL-licensed code. Citadel is an all-in-one messaging and collaboration server that runs on Linux, BSD, Solaris, and pretty much any Unix-like operating system. It includes email, group scheduling and calendaring, shared contacts, mail list server, instant messaging, public folders, and mobile device support. Citadel supports distributed installation across multiple servers for scaling and high availability, Web access and local clients.

The key to understanding how Citadel works is grasping the old Bulletin Board Services (BBS) concept of rooms. A room is any Citadel element: email folder, RSS feed, calendar, chat—anything at all. It's an easy installation, since it is included in most Linux distributions, and it's easy to administer—you can literally be up and running in an hour.

There is good community support for Citadel, and there is just one version—their best one.


3. Open-Xchange

Open-Xchange server serves up several versions: the free Community edition, the Advanced Server edition and the Hosting and Server version. The Community edition is not supported by Open-Xchange, and doesn't include proprietary modules like OXtender for MS Outlook and OXtender for Mac OS. If you want those modules, then you have to buy one of the other editions. Otherwise it's similar to the commercial versions.

The Advanced Server is a complete system that includes the Debian Linux-based Univention Corporate Server operating system, and a raft of nifty tools such as the Open-Xchange Microsoft Outlook Uploader (for migrating data from Outlook or an MS Exchange server), the OXtender for Mobile Web, and an MS Outlook updater. If you have a lot of Microsoft clients, or you need to interoperate in a Microsoft network, the Advanced Server makes it fairly painless.

The Hosting Edition is for ISPs and hosting services that want to offer hosted email and groupware services.

A new feature is the OX App Suite, a messaging suite that includes a word processor, spreadsheet, calendar, contacts, and instant messaging and sharing.

Open-Xchange is a member of Lisog, the open source stack initiative, with the goals of helping businesses craft open source strategies, and integrating propriety and open source software.


4. Zimbra

The Zimbra collaboration suite is another commercial many-bells-and-whistles, open source-based server with proprietary add-ons. It includes all the usual goodies: email, Webmail, shared folders, shared contacts, calendaring and scheduling, instant messaging and mobile apps, and Outlook sync.

Zimbra's Web interface is called Zimbra Desktop. Zimbra Desktop supports all of Zimbra's features, and it supports syncing external accounts such as Gmail, Yahoo, or any POP3 or IMAP mail. In fact, a feature common to all of the groupware suites in this article is they have excellent Web-based client interfaces, so you don't need standalone clients like Outlook or Evolution. If you can wean users away from their beloved Outlook you can make administration a lot easier, and not have to hassle with the cost of using special Outlook connectors.

Zimbra has Zimlets, which is their name for mashups, which is another name for usefully-collating multiple sources of information. This lets you do things like check your calendar, set appointments, and pick a restaurant to meet in without ever leaving the email you're reading. Zimlets make a lot of information available on mouseovers such as phone numbers, flight schedules, status messages, maps, address books entries, and lots more. (A picture is worth a thousand words, so check out the demonstration video.) You can choose from a whole gallery of Zimlets, and you can write and share your own.


5. SOGo

The historical stumbling block for replacing MS Exchange has been Microsoft's closed, proprietary MAPI protocol. In 2007 the European Union ordered Microsoft to open several of its protocols such as CIFS (Common Internet File System), MS Active Directory protocols and MAPI. When Samba 4 is released it will have full support for the Microsoft protocols, providing an Active Directory alternative. MAPI is key to natively supporting MS Outlook, and to interoperability between Exchange and other groupware and mail servers.

The OpenChange project has been developing a portable open source implementation of MAPI, with the goal of native Exchange and Outlook support, and you can see this in action in the SOGo groupware suite. SOGo supports multiple languages, and multiple clients such as Outlook, Thunderbird, Blackberry, iPhone, and Android. SOGo is free software and free of cost.



Zarafa, based in the Netherlands, also offers native Outlook support. Zarafa is designed to add on to your existing mail server and WebDAV server. An easy way to get acquainted is to set up a Fedora Linux test server, because Zarafa is included in Fedora.

Read more »

Faster Software through Vectorization and Parallelization
Posted by Thang Le Toan on 24 May 2016 05:15 AM

How can developers continue to achieve performance gains? Here’s a look at how to take advantage of the cores at your disposal

Everyone agrees there is a need for speed, and pressure is on developers and engineers to deliver better performance from every new piece of hardware. In this post, Slashdot Media Contributing Editor Rick Leinecker looks at how Vectorization and Parallelization help achieve that.

Faster Software, Faster
Rick Leinecker, May 2016
Everyone loves programs that run fast and smooth. For instance, Microsoft Word does complex image manipulation easily without noticeable delay. The march of software toward even greater levels of performance helps satisfy our need for speed.
The rule of thumb for gains in computer performance can be summed up with Moore’s law. It states that every two years, computer hardware doubles in capability, which is also accompanied by price reductions. This explains the aforementioned march to greater levels of performance along with the widespread hardware bargains. But computer clock speeds haven’t improved much since around 2005, while software still continues to enjoy ever-improving performance.
In the absence of improved computer clock speeds, Intel provides ways for developers to continue creating faster and faster software. I am going to mention the two methods that have provided my software development the biggest gains: vectorization and parallelization.
To explain vectorization, I will use an illustrative metaphor. Let’s say that you own a store with a long rack filled with claw hammers. Your accountant sent an email stating that you are losing money on the hammers, and you need to raise the price. Still using old fashioned labels, you figure out how long it will take your stock man to change the labels. It will take hours with the label machine that you own. But a slick salesman walks in at that opportune moment and says he has a new-fangled labeler that can affix labels on four items at a time. This is good news. Your stock man can do the job in a quarter of the time you originally estimated. He takes the new labeler down the aisle and does the task, four hammers at a time.
This is similar to how Intel CPUs do vectorization. In many situations, the CPU provides instructions that operate on multiple adjacent items, similar to the new-fangled labeler. The technology is known as Single Instruction Multiple Data (SIMD). The new CPU instructions can perform operations on anywhere from two large pieces of data to sixteen small pieces of data. For instance, if the program needs to add a value of 5 to a list, it can add 5 to multiple list items simultaneously. For garden variety numbers, this might give you a speed up factor of 4. Figure 1 illustrates how CPU single operations affect multiple data elements.


Figure 1: Single CPU operations affect multiple data elements.
In order to employ vectorization via the SIMD technology, all developers must do is compile their software with a modern compiler. Modern compilers inject the special SIMD CPU instructions on your behalf at compile time—it usually requires no effort on your part except for the recompile. I could take an old piece of software off the shelf from ten years ago, recompile with the Intel compiler, and for code that acts on adjacent data enjoy a huge performance gain. In my day-to-day programming I normally get a four-fold performance boost where vectorization is applied.
As elegant as the new-fangled labeler was, there was another option for the relabeling job. You could have asked another three employees to help, for a total of four to perform the task. Then, instead of a labeler that can re-label four hammers at a time, you now have four people to re-label the hammers with the old-style labeler. Each person works on a separate section of the hammer inventory to complete the task.
This is similar to parallelization. Software has constructs called loops that perform sequential groups of tasks. These loops can be split up to take advantage of the cores on a system, similar to the way the four employees might share the labeling job. For instance, if you have a four core CPU, then loops can be split up four ways. Each of the four cores takes a quarter of the loop’s task, thus finishing the task four times as fast. And if you have more than four cores, the loop’s task can be split up into even smaller sections, thus improving performance that much more. While the compiler does the work in splitting up loops behind the scenes by injecting the appropriate code, the developer must direct the compiler to do so before compilation with special directives. Figure 2 illustrates how this idea works.


Figure 2: Loop parallelization uses four distinct processes to split up a sequential task.
Now here is where it gets really interesting. Let’s say you are back to engineering the re-label job. What if you have four of the new label machines with four of your employees to do the job? Now you can go sixteen times faster and get the job done quickly. Each of the four employees is re-labeling four hammers at a time. The same can be true for software. Compilers can use both vectorization and parallelization for dramatic performance gains. This is the beauty of Intel’s CPU technology and modern compilers which target them. Even without improvements in CPU clock speed, your software can still run faster. The combination of vectorization and parallelization is a winning combination, and software is the beneficiary.
Finally, I’d like to point out that combining both techniques can actually yield better than a 16 time benefit. Let’s say that you have a computer with eight cores. Now let’s say that you are acting on a list of single bytes, which can be acted on 16 at a time. This would yield a performance increase of 128 times. And if this is older software that you took off the shelf for a recompile, that’s not bad for a relatively small effort.

Read more »

Help Desk Software by Kayako