Nice video krub, i want to know about alternative of "consistency hashing" please leave keyword again?
@BBConfigАй бұрын
Thanks very much krub. The alternative is Rendezvous Hashing which is stateless, doesn't need all the complication of virtual node - blog.techlanika.com/rendezvous-hashing-dad794931065
Art work ในวิดีโอ จาก instagram.com/gulin.watercolor/?hl=en ครับ
@dream073122 ай бұрын
Thx a lot nice video krub, i have a question, How do we know if the bloomfilter space initially set is no longer enough?
@BBConfigАй бұрын
Thanks very much krub. Most of the time, bloom-filter is used in system that can (more or less) control the expect entries of data krub. In use case of my project, we limit maximum data entries in a chunk of data (1 chunk for one bloom instance). So we let library calculate the size of bloom e.g newBloom(maxNumEntries, false-positive-rate) . So we don't really monitor if false-positive rate is increase or not, we just control the max-entries inserted to the bloom instead.
@dream07312Ай бұрын
@@BBConfig Thank for your answer krub 🙏
@BBConfig2 ай бұрын
ข้อมูลเพิ่มเติมที่ผมเพิ่งเห็น ตัว doc เองก็พูดถึงเคสที่ ReadWriteLock อาจจะไม่ช่วยให้ performance ดีขึ้น " if the read operations are too short the overhead of the read-write lock implementation (which is inherently more complex than a mutual exclusion lock) can dominate the execution cost" docs.oracle.com/javase/7/docs/api/java/util/concurrent/locks/ReadWriteLock.html วิดีโอต้นเรื่องที่ผมไปเจอที่อ้างว่า ReadWriteLock นั้น overhead สูง kzbin.info/www/bejne/pmW3oqaamK-td8k
@BBConfig2 ай бұрын
"Confluent Cloud Is Now 100% KRaft and You Should Be Too" - www.confluent.io/blog/zookeeper-to-kraft-with-confluent-kubernetes/
@BBConfig2 ай бұрын
Data ที่ผมใช้มากจาก Kaggle : Goodreads Book Reviews Data Set - www.kaggle.com/datasets/pypiahmad/goodreads-book-reviews1?select=goodreads_books.json