S3 retrieval anomaly

[M1Systems]

Why would I by getting huge spikes for retrieval from S3 when I haven’t been run any retrieval jobs?