6.4.1.1 Data Log Trend recording

The actual number of tags that can be Data Logged by a SCADA node is dependent on:

· Processor Load (due to other tasks like scripts, schedules, communications, calculation tags).

A guideline is 500 tags/second for a 1.8 GHz Pentium IV processor using an IDE disk controller. If your Scan rate is every 3 seconds, then you could data log about 1500 tags. Increasing the deadband to reduce the number of changes/second could increase this number of tags data logged. These assume the worse case scenario that there is no deadband or that the tags exceed the deadband every scan cycle.

The use of a deadband will result in recording only significant changes, effectively reducing the number of value changes recorded per scan (and increasing the number of Tags that can be data logged).

It is recommended to use SCSII drives for data logging. A separate Data Log drive can be specified under SCADA node properties Data Log Folder.

Data Log Trend Disk Space

Log to ODBC (Analog Tag Log, Discrete Tag Log, System Log, Action Log and Alarm Log) record data to an ODBC Database on the hard drive of the Project Node (an Access Database by default, named BwPData.mdb via the DSN named bwpdata_Access)

Data Log Trend records 10 bytes per sample on the SCADA node. If you are sampling 1000 tags every 1-second and you are using NO deadband, then it will be:

500 tags * 10 bytes/sample * 1 scan/ sec * 3600 samples/hour * 24 hours/day * 30 days/month * 1/1024 *1/1024 *1/1024 = 12 Gigabytes in a month.

500 tags * 10 bytes/sample * 1 scan/ sec * 3600 samples/hour * 24 hours/day * 365 days/year * 1/1024 *1/1024 *1/1024 = 146.9 Gigabytes in a year.

500 tags * 10 bytes/sample * 1 scan/ sec * 3600 samples/hour * 24 hours/day * 1/1024 *1/1024 = 412 Megabytes in a day.

If a deadband is used, it will be less. If you assume the tags exceed deadband only 50 percent of the time (every 2 seconds), it will be

500 tags * 10 bytes/sample * 1 scan/ 5 sec * 3600 samples/hour * 24 hours/day * 365 days/year * 1/1024 *1/1024 *1/1024 = 29.4 Gigabytes in a year.

The Log Data Maintenance feature in Advantech WebAccess/SCADA will archive Data Log Trend files to a network folder or mass storage device AND will ERASE files that are older than a user-defined period on the local hard drive of the SCADA Node. Log Data Maintenance will also archive and delete expired (old) records from the ODBC Log databases on the Project Node. (See Log Data Maintenance for more information). The Log Data Maintenance will prevent your disk drive from filling up if used properly. In the above examples, if Log Data Maintenance were set for 30 Days would require only 9 Gigabytes of Disk space for data log trend files for 500 tags.

Note - if both Archive and Delete Expired Files are selected, but the Archive device is off-line, then the expired records and files will not be deleted.

Archive / Mass Storage Disk Space

Data Log Maintenance will copy expired Data Log Trend files daily to the archive media. No file compression is used in version 3.0. In the above 500 tag examples:

412 Mega Bytes /day / 500 tags / sec assuming every tag changes every second with no deadband for Data Log Trend files.

Data Log Maintenance will also create Access Database files daily for expired records from the ODBC Logs; up to 6 files daily: System Log, Action Log, Alarm Log, Analog Tag Log, Discrete Tag Log and Text Tag Log, as selected by user. If the SCADA node is off line, then a data file will be created the next day the SCADA node is running at the record time, containing data from the previous day(s).

The minimum size is 64 Kbytes per daily ODBC file. (6 x 64Kbytes = 384 Kbytes / day minimum). The maximum is hard to calculate for System, Alarm and Action Log.

Analog Tag Log (bwAnalogTable - bwPData.mdb) 29,900 Kbytes / 370,000 Value changes

Discrete Tag Log (bwDiscreteTable - bwPData.mdb) 14,200 Kbytes / 250,000 State changes.

There is no Archive restore or Playback feature in version 4.0. An Archive restore and playback is scheduled for a future release. Users can manually copy Data Log trend files (second X, minute M and hour H files) back to the SCADA node to view archived data. Archived ODBC files are easily opened using Access.