AMBITION: Ambient Temperature Aware VM Allocation for Edge Data Centers

Edge data centers are increasingly deployed to improve response time of intelligent services. Due the high computing demands for such services, edge consume a considerable amount power, generating excessive heat. To mitigate thermal problems with smaller cooling usually trigger software-based management techniques along air systems. Unfortunately, ambient temperature servers often has surge due consolidation VMs and heat propagation among components (e.g., CPU, GPU, memory unit, disk, etc.). Higher further increases on-chip temperature, invoking more frequent throttling. resolve deteriorated by in this paper, we propose an aware VM allocation technique, called AMBITION. Considering performance impact AMBITION estimates actual capacity servers. Based on VMs, finds appropriate server which sufficient ambient-aware run VM; it allocates computation-intensive higher capacity, distributes memory-intensive individual as much possible. In our experiments center, shows execution speedup 50.3%, average (up 73.8%), compared conventional technique while saving system-wide energy 5.9% 13.6%). At expense 5.8% (from 50.3% 44.5%), saves power 84.3%, leading 29.3% total center saving.