runtime • It dynamically routes data after measuring which ordering is the most profitable Dynamic re-ordering with Eddy B A D C Eddy C D A B ??? Vasiliki Kalavri | Boston University 2020 18 Data-parallel streaming languages enable fission by construction • Elastic scaling techniques enable dynamic operator fission by adjusting the number of parallel operator instances according to data rates constraints: what are the trusted hosts for each operator? • Ensure state migration: if placement is dynamic and the operator is stateful, its state must be moved in a consistent manner Operator placement

0, we receive one event: • Insert-only edge stream: events indicate edge additions • Fully-dynamic edge stream: events indicate edge additions or deletions A t+1, the graph is obtained by inserting nton.pdf • Stefani, Lorenzo De, et al. Triest: Counting local and global triangles in fully dynamic streams with fixed memory size. TKDD 2017. https://www.kdd.org/ kdd2016/papers/files/rfp0465-de-stefaniA

Stream Models Vasiliki Kalavri | Boston University 2020 A stream can be viewed as a massive, dynamic, one-dimensional vector A[1…N]. The size N of the streaming vector is defined as the product of negative. Events can be continuously inserted and deleted from the stream. It can model fully dynamic situations: • Monitoring active IP network connections is a Turnstile stream, as connections can

data when input rates increase beyond system capacity. • Load shedding techniques operate in a dynamic fashion: the system detects an overload situation during runtime and selectively drops tuples

load is Θ(ln n/ln ln n), with high probability ??? Vasiliki Kalavri | Boston University 2020 Dynamic resource allocation • Choose one among n workers • check the load of each worker and send the