nsq and scopedb resource issue under high write load

We have two problems, Asim. First thing is that as to switch that in OAMP, once we enable double-write on Kudo and KudoX or ScopeDB writing, once we fix that by changing the parameter in terms of re-queue on KudoX, you have another problem coming two days later, where there was just not enough processing power by KudoX and once inserted, so the SQ keep piling, until a point where the disk didn't have enough space for all the piled messages in SQ, and it just failed overnight. We fix that by increasing the storage of the disk and increasing the processing power of KudoX, which is, for example, scaling number of costs, and then we increased the concurrency of Gwensel Acer by changing the file to starting argument.