EpicorFree calculatorPDF, Excel, Word exportEpicor KineticERP Implementation

Epicor DMT Load Duration Estimator

Cutover weekends fail when the data load takes twice as long as planned. This estimator turns record counts and a measured load rate into a realistic window, including the error reloads and per-file checking that people forget. Use it to decide how many rehearsals you need and whether the go-live window is long enough.

Your numbers

records

Sum across parts, BOMs, customers, suppliers, open orders and balances.

records/min

Time a sample load on the target environment. Rates differ a lot between tables.

files
min

Open, validate, review log and check counts.

%

Rows that fail and must be corrected and reloaded.

min
runs

Your results

Time for one full load cycle
546.3 min
Cutover window needed
11 hrs
One load cycle plus contingency.
Raw load time
375 min
Error reload time
11.3 min
File overhead
160 min
Total rehearsal effort
27 hrs

A planning aid only. Always time real loads on your own environment. Dependencies between tables can force serial loading.

Save results:

Load cycle time vs total records

Sweeps Total records to load from 10000 to 500000, holding your other inputs fixed.

Migration plan review

Optional: have an expert review your data scope, sequence and cutover timings.

Optional. The tool and every export work without this. No spam.

How to use it

Time a sample load of each major DMT file in the target environment, then use a blended records per minute figure. Enter the total record count, the number of files, and the minutes your team needs to open, validate and check each one. Add the error rate seen in rehearsal and a buffer. The tool returns the time for one full cycle and the window to book. Rerun it with the rate from each rehearsal; it should converge.

  • Rates vary by table: BOMs and operations are slower than simple masters
  • Load sequence matters, parts before BOMs, customers before orders
  • Never assume the production rate equals the test rate

The estimation logic

The total is the raw load time, plus the time to reload the rows that fail, plus a fixed overhead for each file. The chart is linear in records, which is the main insight: doubling volume doubles the cycle. If the window is too short, the levers are to cut the data scope (open items only, not history), parallelize independent loads, or improve data quality to reduce reloads.

Reading the output

In the default example, raw loading is about six hours, with another 11 minutes for reloads and nearly three hours of per-file overhead, so one cycle is around nine hours. Add the buffer and you need roughly eleven hours. Notice that overhead is nearly a third of the cycle: automating validation checks is often a better investment than a faster server.

Common mistakes

Teams forget that a rehearsal must be run end to end, including reconciliation, or that a fix to one file forces a reload of dependents. They also load history nobody uses. Agree what data is truly needed on day one, load balances rather than transactions where accounting allows, and keep a signed-off reconciliation list so the cutover has clear exit criteria.

Frequently Asked Questions

How do I measure the load rate?

Load a representative sample, perhaps a few thousand rows of each major file, into the target environment and divide records by elapsed minutes. Repeat at a quiet time of day, as network and database load can change the number.

What error rate should I expect?

Early rehearsals can show double-digit rates; by the final rehearsal it should fall to low single digits. A high rate on the final run means data cleansing, not tooling, is the problem.

Can DMT loads run in parallel?

Independent files can often be loaded at the same time, but dependent ones cannot. Check the load order, and remember that parallel runs may slow each other down, so test the combined rate.

How many rehearsals do I need?

Most projects plan at least two or three full rehearsals. The last should run to the real timetable with the people who will do the cutover, and should be signed off against reconciliation checks.

If you are planning an Epicor data migration and want a second pair of eyes on scope and sequencing, we can review the plan with your team.