This computing proposal meet the needs of a metagenomic project, on the analysis of ancient DNA biomolecules coming from diverse sediments. From Sweden lake cores to Greenland sediments, including other samples like North Sea core, we retrieved metagenomic data through shotgun sequencing in order to get the widest snapshot of former environments. Dealing with such data generated by next generation sequencing (NGS) platforms induced the processing of quite heavy raw sequencing files and specific pipeline and tools related to taxonomic classification. One important aspect of the analysis is notably to been able to differentiate false positive and carefully authenticate ancient taxa. For that, quite massive database are required for mapping or classifiers tools, requiring considerable core hours. We have designed a pipeline, based on different modules installed in Dardel to perform such complex analysis, nevertheless the compute project I belong is most of the time exceeding the monthly allocated core hours, slowing down massively the processing of our sediment samples through this pipeline. In the light of this, I believe that a small compute project would be of great help and would definitely increase the efficiency of ancient sedimentary DNA projects that I'm running during my post-doctoral research at the Centre for Palaeogenetics, at Stockholm University.