Two reasons.

 First, have you ever tried to restore a tape backup file to your server? Chances are this backup system was set up, and is periodically tested and maintained – hopefully! – by your IT person or firm. Chances are also that you don’t really have a clue how to restore a file yourself, or an entire directory, from your backup media, thus requiring you to chase down your IT support resource in the event you need to restore your Peachtree data. By the time that occurs you have either been “down” for longer than you can tolerate, or additional accounting data has been entered into the system, meaning more redundant entry when the database is finally restored.

I still recount a “horror story” from over fifteen years ago, when a client dutifully changed backup tapes every morning, took them off-site every night, and never made an application-based backup. One day they corrupted their database and needed to restore from tape, only to find that the tapes were all blank. No one had ever verified what was being backed up, or tested the restore process to insure its reliability.

 Application-based backups should not be considered as a replacement for an unattended server-based backup system, but as an adjunct to such a system. Server-based backups typically write their data to some type of external storage device, so that if the server drive fails (the most common type of server failure, by the way), your precious data is available to you “detached” from the server drive. It is, however, so much easier to restore an application-based backup file, since the program takes all of the “techno-babble” out of the restore process and guides you through the restore right from within the application.

These types of backups can of course also be stored on external media (thumb drives, USB drives, etc.) but will run much more slowly if directly backed up to an external device. Better to first back up to a local drive, then drag the completed application-based backup file to the external media for secure storage.

 Second, do you really remember – or have the time – to manually run a backup every day? Can you routinely get everybody out of the system, on your schedule, so that you can run the backup? Probably the answer to one or both of these questions is “no”. Peachtree’s automatic backup configuration can eliminate these roadblocks to protecting your valuable accounting data.

 To use the new Peachtree automatic backup feature you run a small “utility” (from within the application, off of the File menu) to specify:

• what company to back up,
• where to store the backup file,
• what credentials you have to permit access to the company database,
• what additional components you may want the backup to include,
• and what to do in the event that a backup file with the same name already exists.

You save these settings in a “configuration file” (one for each company whose backups you want to automate), and then invoke the Windows Scheduler application, which is typically installed on all Windows-based PC’s.

 Windows Scheduler then takes over the process, asking you how often the backup should run, and what time it should kick off. The “no-brainer” setting for this are “daily, in the middle of the night sometime”.  Just be sure that it can complete its backup before your server based-unattended backup starts up, as the one could conflict with the other and prevent a clean backup from either source.

 If you are not yet running on Peachtree 2010, you can still use Windows Scheduler, in conjunction with some old-fashioned “batch files”, to more or less accomplish the same automated backup routine.  Not a simple task, however. You can start a fire with a flint and some straw, as well, but lighters are a whole lot easier. And “easier” usually translates into actually getting the job done reliably and consistently.

When it comes to backing up accounting data, reliability and consistency are the values you are looking for.

 Give us a shout if you would like to learn more about automated backups for Peachtree.

 So, what is this mysterious critter that sits lurking quietly beneath the surface of Peachtree?

 In overly simple terms, it is what is called a “database manager”, storing the data that users enter into Peachtree and “managing” the process of “locating and serving” the right information to users based on their activities in Peachtree (and their security settings). Pervasive manages the byzantine process of updating the Journal Header, the Journal Row, the Customer, and the Item tables every time a user enters an invoice in Peachtree, making sure that all the data goes where it belongs and is properly indexed so it can be located again when needed.

 Pervasive SQL, unlike its more well-known “big brothers” such as Microsoft SQL, Oracle, and MySQL, is designed specifically as a “low cost of ownership” self-administering database manager, with no dedicated Database Administrator (DBA) required, and no real technical knowledge needed to initially install and configure the product. In fact, the Peachtree installation routine makes all of the Pervasive installation and configuration settings transparent. This is a good thing, to a point, and fits nicely with the pricing model and target audience for the Peachtree application.

 However, veteran computer users should recognize that there is no such thing as “one size fits all” when it comes to configuring a database. Some Peachtree users hit the database very lightly, while others process thousands of transactions in many different tables within the database. Hardware and other “environmental” configurations can be vastly different from company to company, which can have implications for performance, and therefore for database configuration. Sage has to account for all possible variations in uses when setting up default configuration settings.

 Because Sage Software has no way of knowing in advance exactly how you plan to use your Peachtree system, or what your transaction volume will be, the only logical response from Sage for initial database configuration is to make the settings conform to a “lowest common denominator”, which addresses the published minimum hardware and overall “technology environment” requirements. That is really the only way to go to insure that the product will work “out of the box.” Keep in mind, however, that the Pervasive SQL engine can only work with what is made available to it, via its configuration settings, and the minimum, automatically-set system configuration does not necessarily insure an optimum setup for your specific accounting software needs.

 So, what is a user to do to wring a bit more performance out of this powerful-but-low-maintenance-oriented database manager?

 There are four factors to consider in “tuning” your Peachtree setup. Three of these are relatively straightforward, are independent of the Pervasive database itself, and can be addressed by most competent IT professionals. The fourth factor involves “performance tuning” for the actual Pervasive PSQL engine on the server, and is a little more involved (including modifying Registry settings on the server, which makes some folks get sweaty palms right away – for good reason).

Server Hardware
Don’t have a dedicated server? Get one. Peachtree doesn’t need a dedicated server to operate, but if the performance of your accounting system is a concern of yours, start here. Using a so-called “peer to peer” network for running an accounting product requires acceptance of lots of performance concessions, which most users grow tired of pretty quickly. Servers today are more reasonably priced than they have ever been.

Look for at least a dual-core processor, around 3 Gigahertz or better. Pervasive does not take advantage of dual processors yet, but if you run other applications from the same server (as most small businesses do), using dual processors will allow the operating system itself to manage some of the “load balancing” between the various apps.

Purchase the fastest disk drives you can afford. Disk access time can make a huge performance difference for a heavily input-output intensive application such as accounting. A “Raid 0” stripe array, while pricey, will allow much faster disk writes than a more “conventional” disk setup. Your money or your time…..

Server Memory
Server RAM is not terribly expensive, and is relatively simple to install these days. The question of “how much” RAM to install can be complicated, and the answer is generally dependent of what apps will run on the server. Of course no one ever has good answers to all of these questions when purchasing a new server, so there are two rules of thumb here.

1. Never buy less than 4 Gb of RAM for a modern server.
2. Always ask your server vendor about memory “deals” (2-for-1, additional discounts for larger chipsets, etc. etc.) and then purchase the most memory you can afford within the confines of  that “deal”.

Windows servers have some limitation on how much memory a user can address, but having plenty of memory will insure that there is always more than enough left for other server purposes after the user has maxed out the user-addressable memory available.

Network Environment
Many apparent database performance issues are actually a consequence of the network “infrastructure”, and these can really be the devil to diagnose. There are a few obvious things to consider here.

1. Network interface cards (NICs), routers, and network switches are all critical components of the computing environment, and it rarely pays to “skimp” on those components. Purchase high quality components from known manufacturers, and have them configured by knowledgeable IT professionals.
2. Don’t have your network cabling installed by your brother’s nephew (unless he is a competent IT guy).
3. Don’t use a wi-fi connection for a large database; wi-fi is neither fast enough nor reliable enough (with microwave ovens, cell phones, etc. interfering) to trust for accounting transactions.

 Configuring the Peachtree Pervasive Database Engine
This is the fourth and final piece to focus on to “tune” your Peachtree install for optimum performance. I will address this final piece of the puzzle in my next post on PeachFuzz.

1.  Tweak your anti-virus settings – Most anti-virus software allows you to define “exclusions” to virus checking. You should “exclude” the folder where you installed Peachtree on your local drive, as well as the folder and all subfolders on the server where your Peachtree data is stored. This will not likely put your computer in harm’s way; we have not heard of any malicious software ever infecting a Peachtree .DAT file.
2. Free up local hard drive space – You should try to make sure that you have three times the size of your largest company database or 1 Gb, whichever is greater, available as free space on your local hard drive. If Peachtree needs to create a Windows “swap” file, this is where it will be parked. Running a “disk defrag” on the local drive to optimize the available storage space can also help in some situations.
3. Edit your Peachtree global settings – There is a section on the “General” tab under the Global Settings menu that allows you to set up “smart data entry” rules. While enabling these settings can be handy and make the system more convenient to use, they can also contribute negatively to performance. Unchecking all of the checkboxes in this area may improve speed.
4. Disable Events- This is an area of Peachtree that we find hardly anyone uses, but it may still be “activated” on your system. To turn this off (which can improve performance) click on Tasks, then Action Items, then click on the “Options” button on the Action Items menu bar. From the Options menu select the Transactions tab, and uncheck any of the boxes that are checked off in the “Create Event” column.
5. Avoid duplicate transaction reference numbers – Peachtree allows you to use the same “reference number” (e.g., vendor invoice #, cash receipt ID, etc.) as long as you do not use it for the same “master record” (vendor, customer, etc.). However, using the same number over and over (e.g., “cash” for all deposits) causes the application to work harder when searching and storing information. Keeping all of your “reference numbers” unique to the greatest extent possible can improve performance. It can also make life much more pleasant in the unhappy event that you ever need to export and re-import your data; the import will sometimes reject duplicate reference numbers.

 Of course the most significant contributor to system performance is the size of your Peachtree database, and within that database the single biggest contributor is the Journal Row (JRNLROW.DAT) table within the database. This table holds ALL of your Peachtree transactions, and each time you search for a transaction record the system needs to read through that entire table to find the requested record.

 So, how big can this file safely grow to before performance becomes an issue? There is no hard and fast rule for this (still more art than science), but our general “rule of thumb” is to begin thinking about maintenance to your database when the journal row table approaches 150,000 total records, if you are using any version of Peachtree except Quantum. The more powerful Quantum product can gracefully accommodate three to four times that number of records without causing significant performance “degradation”.

 You can review the record count for all of your Peachtree tables by selecting Help | Customer Support and Service | File Statistics from the top menu bar within Peachtree.

 Once you have determined the overall size of your Peachtree database and viewed the record counts for the individual tables, you can make a judgment about purging some data. Purging is the process of physically removing selected records from the system, to shrink the overall size of the database. It can, however,  be a time consuming and technically daunting task, since there are any number of “gotchas” that can impede or abort the purge process.

We have assembled a set of instructions for performing Peachtree database purges on our website, which you can review by clicking here. In some circumstances it is more cost-effective to use an outside data repair service to fully purge records that are no longer needed in the system. And in all cases, it is good practice to make an “archive” copy of the database before starting a purge. Peachtree Premium makes the archiving process easy, but Peachtree Complete users can create “manual” database archives.

 Carefully following these “best practices” for Peachtree data storage will help you to get maximum performance from your Peachtree software.

However, typing an entry into one of these “drop down” fields will add that entry to the drop down list automatically if it is not an entry already on the list. Most of the time this is what you want to do, but sometimes data gets into these fields that is clearly incorrect and doesn’t belong there. Problem is, once it’s on the list anybody can select it on a subsequent record entry and compound the error.

So, how do you get rid of these unwanted entries?

Cleaning this up is a two-step process, First, find all of the records that have the incorrect entry and edit all of those records to replace the field(s) with the correct entry; this can be done manually, or via a mass export/import (call us if you don’t know how to do this).

Next, once you are sure that no records are holding the values you want to remove, run a Peachtree Data Verification (File | Data Verification). Choose the option to run both tests, and choose to make a backup before running the tests (always back up before running any database maintenance task).

Under the Hood: The Data Verification process checks the values in the AFLOCAT.DAT table against all the values in all the records, and removes any values that are unused (e.g, mis-typed, mis-”cased”, etc.).

Once complete, those unwanted, unneeded choices will no longer appear in your drop-down lists.

Why should I bother to upgrade, my software runs just fine?  We offer two responses to clients who offer this all-too-familiar refrain.

First, Sage Software’s support policies provide end-user technical support for the “current” version of Peachtree and the prior two releases. Thus with the release of Peachtree 2010, Sage will shortly discontinue product support for any versions of Peachtree older than the 2008 edition. There are obvious revenue generation motivations for Sage to follow this course, but it also speaks to the difficulty of maintaining a support organization with expertise in many different product editions.  Software telephone support is not a long-term career choice for most aspiring technologists, so these positions inevitably turn over pretty rapidly. Chances are, after three years there are very few support technicians at Sage who have been exposed to a version of Peachtree that is more than three years old.  TriStar’s staff does not turn over, so we support any version of Peachtree in use, but our options are similarly limited if we need to turn to the “manufacturer” for product support on your behalf.

Second, Peachtree users who process payroll within Peachtree absolutely need to remain on a supported version of the product. When payroll tax tables change, payroll tax forms change, or more significantly, when payroll tax policy changes (forcing a change in the actual processing instructions) Sage will only make those changes to the product editions presently supported.  With health care reform, economic stimulus efforts, and a myriad of other “sea changes” emanating from Washington, it is very likely that “older” versions of the Peachtree software will not be capable of processing these changes.

So, the practical implications of this policy is that you need to upgrade your system no less frequently than every three years.  With the “life cycle” of computer hardware now acknowledged to be less than three years, there are often “technological imperatives” for upgrading your accounting software. Applications released more than three years ago (and likely designed well before that) are unlikely to behave flawlessly in a new operating system, on new computer hardware, that didn’t exist when the accounting application was originally designed.

Love it or hate it, responsible business owners should expect to upgrade their “business critical” software applications (of which accounting is certainly one) no less frequently than every three years.

So, how old is your accounting software?