Search for: All records

Distributed caches are widely deployed to serve social networks and web applications at billion-user scales. This paper presents Cache-on-Track (CoT), a decentralized, elastic, and predictive caching framework for cloud environments. CoT proposes a new cache replacement policy specifically tailored for small front-end caches that serve skewed workloads with small update percentage. Small front-end caches are mainly used to mitigate the load-imbalance across servers in the distributed caching layer. Front-end servers use a heavy hitter tracking algorithm to continuously track the top-k hot keys. CoT dynamically caches the top-C hot keys out of the tracked keys. CoT’s main advantage over other replacement policies is its ability to dynamically adapt its tracker and cache sizes in response to workload distribution changes. Our experiments show that CoT’s replacement policy consistently outperforms the hit-rates of LRU, LFU, and ARC for the same cache size on different skewed workloads. Also, CoT slightly outperforms the hit-rate of LRU-2 when both policies are configured with the same tracking (history) size. CoT achieves server size load-balance with 50% to 93.75% less front-end cache in comparison to other replacement policies. Finally, experiments show that CoT’s resizing algorithm successfully auto-configures the tracker and cache sizes to achieve back-end load-balance in the presence of workload distribution changes. 

Bitcoin [12] is a successful and interesting example of a global scale peer-to-peer cryptocurrency that integrates many techniques and protocols from cryptography, distributed systems, and databases. The main underlying data structure is blockchain, a scalable fully replicated structure that is shared among all participants and guarantees a consistent view of all user transactions by all participants in the system. In a blockchain, nodes agree on their shared states across a large network of untrusted participants. Although originally devised for cryptocurrencies, recent systems exploit its many unique features such as transparency, provenance, fault tolerance, and authenticity to support a wide range of distributed applications. Bitcoin and other cryptocurrencies use permissionless blockchains. In a permissionless blockchain, the network is public, and anyone can participate without a specific identity. Many other distributed applications, such as supply chain management and healthcare, are deployed on permissioned blockchains consisting of a set of known, identified nodes that still might not fully trust each other. This paper illustrates some of the main challenges and opportunities from a database perspective in the many novel and interesting application domains of blockchains. These opportunities are illustrated using various examples from recent research in both permissionless and permissioned blockchains. Two main themes unite the various examples: (1) the important role of distribution and consensus in managing large scale systems and (2) the need to tolerate malicious failures. The advent of cloud computing and large data centers shifted large scale data management infrastructures from centralized databases to distributed systems. One of the main challenges in designing distributed systems is the need for fault-tolerance. Cloud-based systems typically assume trusted infrastructures, since data centers are owned by the enterprises managing the data, and hence the design typically only assumes and tolerates crash failures. The advent of blockchain and the underlying premise that copies of the blockchain are distributed among untrusted entities has shifted the focus of fault-tolerance from tolerating crash failures to tolerating malicious failures. These interesting and challenging settings pose great opportunities for database researchers.