Samsung 950 Pro
There is a new sheriff in town.  This storage device is faster, much faster, than any storage device ever.  Built using Samsung's V-NAND process, the M.2 form factor drive is able to claim sequential read performance of 2,500MBps, and sequential write performance of 1,500MBps.

We designed a POS solution to record sales and to provide meaningful reports

Touch 'n' View

A Scalable Framework and Interface to Store and Query Transactions in a Hospitality Enterprise

ViewTouch data is always located in main memory (as well as in storage, of course)!  Having ALL the data in main memory as an Integral part of the POS system makes ViewTouch a much faster and superior point of sale solution.  ViewTouch avoids the BIG MISTAKE of recording your transactions in a third-party's 'Commercial Database'.  ViewTouch stores transactions the RIGHT WAY, in a 'Flat File' subsystem which exists entirely BOTH in MAIN MEMORY and in permanent flash storage!  Relational Databases simply can't be located in main memory on POS systems because they're too big.  They're never an integral part of the POS application itself in the way that ViewTouch Flat Files are!
The moment all of the data is available in memory, it is an entirely new game: access time and seek times become irrelevant (no disk seeks).
We'll let someone who earns his living in the Commercial Database business explain why you should NEVER use a Commercial Database (such as SQL Server or Oracle) to record information such as your restaurant's transactions.
Caution: Optional Reading Ahead

You don't really need to understand why our 'Flat File' approach is absolutely the best, right way to record your transactions, but if you do care to know why then you can continue reading.  Feel free to just skip this part and continue On To Table Management if you want to.
The main benefit to you of ViewTouch flat files instead of a database for storing and indexing data is that it increases "Data Availability" and decreases "Cycle-Time", for maximum success. There's no need for monthly data warehouse refreshing. This greatly increases data "freshness" and your ability to make timely decisions. ViewTouch directly accesses the file system without having to build the indexes on the database. This approach is simplest, and minimizes build and refresh times.

Flat files already exist

Typically, data feeding into a data warehouse or database is extracted from some other system in the form of flat files, often from a mainframe. If you don't have to load the information into a database to ensure referential integrity of for some other reason, it is possible to index it directly. ViewTouch has an integral database search engine. Data cannot be deleted from flat files but can be flagged as deleted, which serves the equivalent function.

Less build time and disk space

It takes less time to build indexes from flat files than databases, because you don't need to load the data from some other source into a database. ViewTouch reads flat files directly, without incurring the overhead of the database management system itself. This decreases the "Cycle Time" of loading data and building or re-building the indexes, and increases the "Data Availability" because the data is more available for retrievals. In addition, it takes less disk space to index flat files because ViewTouch stores location pointers instead of row id's with the indexed values.

Quick and easy to manipulate

Data in flat files is very easy to manipulate for prototyping and changing content, or in case a data type needs to be altered or the size of a field/column changed. This is very useful for "exploration" data warehouses, where recursive "design on the fly" and "very ad hoc" query processing are essential.

No temp space issues

When indexing flat files, you need not worry about running out of temp space, as you do with databases. And you still gain the advantage of parallel processing because ViewTouch can build indexes concurrently across multiple servers even with flat files.

No special DBA efforts

When using flat files instead of a database, no special Database Administration knowledge is required in order to create and load the database.

No DBMS license required

When using flat files instead of a database, no licensing fee must be paid.


Flat files are a simpler, easier, faster way to store and access data, if you don't require an actual DBMS for reasons other than data access.

ViewTouch does NOT store transaction data in a relational database management system (RDBMS) based on the relational model. Relational databases grow with every transaction until they become VERY large. The larger they become, the more hardware it takes to deal with them. As they grow they slow down virtually every aspect of your PC's operation. Relational databases are not easily consolidated. If another POS company stores your data in a relational database because the programmers don't know how else they might store your data, then it may well be that the people who implemented the relational database on that POS system don't really know much about storing data, and it is likely that that business's ability to actually derive any value from the data is severely obstructed (and here's why).

We store your transaction data in a format we designed to make it quick & easy for you to learn WHAT you want about your transactions instead of burdening you with having to know HOW to query a database. There is NO need for ViewTouch users to get get lost in the complexities of Sequential Query Language (SQL) server software or hardware, or to learn how to use SQL to try to make use of the vast body of your organization's transactional data and to manage your operation.

ViewTouch transaction data is NOT handled by, managed with or queried by any third-party database software. This data, YOUR data, is TOTALLY secure in this regard. The advantages of avoiding third party software in a POS system may not be obvious, so we'll mention this significant, but typical, example of what invariably can and does happen when your POS vendor relies on third party software.

We created ViewTouch using third generation programming languages but we also created a state of the art, natural-language, fourth generation graphical query language which is built right into the graphical touchscreen interface. We call it Touch 'n' View and you can see it in each of the reports shown above. It's sophisticated & powerful, but it's also intuitive & easy to use. With it, you have the access you need to all of your transaction data. Touch 'n' View makes SQL obsolete, really, and provides you with a much more modern, simpler, intuitive way to create consistently meaningful reports from the transaction records that your business generates.

Valid HTML 4.01!