Performance degradation after some time

Hallo,

I'm having the issue that at a certain point cAlgo becomes slow. I was doing a lot of optimization runs lately and in parallel additional backtesting - assigning 3 cores to optimization and using one for backtesting. Usually optimization uses almost 100% of the cpu when all 4 cores are assigned. At a certain point cAlgo becomes a little bit unresponsive. When I start a new optimization run with all 4 cores assigned, I can then see in the task manager that only about 50%-70% of the CPU is used and not almost 100% as it used to in former optimization runs. I also notice then, that when no optimization or backtesting is running, cAlgo nevertheless uses upto 10%-20% of the cpu. Also (without an optimization running) backtesting two sets at the same time is noticeably slower: 2:15 minute vs 6 minutes for the same two runs, compared with the degraded cAlgo and a fresh opened cAlgo, and the cpu load is ~47% for the degraded and ~57% for fresh started one. Switching between the two running backtests a delay is noticeable in the degraded cAlgo. Memory usage at that time (last case of degraded cAlgo) is several GB, probably due to optimization results and backtest results.

To me it seems that some core component of cAlgo causes a constant workload and so this or another dependent core component is not able to work at full speed anymore, resulting in the described performance degradation to everything else.

Is anyone else experiencing this issue?

Spotware, are you aware of this degradation issue?

My machine: Core I5-2500K, 24 GB RAM, Windows 10 Pro, Samsung Evo 850