-
dbis
-
Publications
-
2012
-
Efficient Frequent Item Counting in Multi-Core Hardware
(P)VLDB 2012 Reviews
are an issue. What applications need 2000 queries/second (Figure 13)? What are these queries? D2: (On W2) Skew in real-life datasets is an established fact. However, the paper makes a leap of faith when [...] that can exploit data skew; thereby also bringing a pipelined parallelism approach for efficiency. S2: The filtering stage is implemented efficiently with vectorized instructions and no branching. S3: The [...] negative aspects W1: The overall problem setting is never made clear. (See detailed description below.) W2: The efficiency of the approach only shows when the degree of skew is high. W3: While the paper focuses …