We describe how we evolve Omid, a transaction processing system for Apache HBase, to power Apache Phoenix, a cloud-grade real-time SQL analytics engine. Omid was originally designed for data processing pipelines at Yahoo, which are, by and large, throughput-oriented monolithic NoSQL applications. Providing a platform to support converged real-time transaction processing and analytics applications - dubbed translytics - introduces new functional and performance requirements. For example, SQL support is key for developer productivity, multi-tenancy is essential for cloud deployment, and latency is cardinal for just-in-time data ingestion and analytics insights. We discuss our efforts to adapt Omid to these new domains, as part of the process of integrating it into Phoenix as the transaction processing backend. A central piece of our work is latency reduction in Omid's protocol, which also improves scalability. Under light load, the new protocol's latency is 4x to 5x smaller than the legacy Omid's, whereas under increased loads it is an order of magnitude faster. We further describe a fast path protocol for single-key transactions, which enables processing them almost as fast as native HBase operations.

As transaction processing services begin to be used in new application domains, low transaction latency becomes an important consideration. Motivated by such use cases we developed Fragola, a highly scalable low-latency and high-throughput transaction processing engine for Apache HBase. Similarly to other modern transaction managers, Fragola provides a variant of generalized snapshot isolation (SI), which scales better than traditional serializability implementations.

Most datacenters still use Equal Cost Multi-Path (ECMP), which performs congestion-oblivious hashing of flows over multiple paths, leading to an uneven distribution of traffic. Alternatives to ECMP come with deployment challenges, as they require either changing the tenant VM network stacks (e.g., MPTCP) or replacing all of the switches (e.g., CONGA). We argue that the hypervisor provides a unique point for implementing load-balancing algorithms that are easy to deploy, while still reacting quickly to congestion. We propose Clove, a scalable load-balancer that (i) runs entirely in the hypervisor, requiring no modifications to tenant VM networking stacks or physical switches, and (ii) works on any topology and adapts quickly to topology changes and traffic shifts. Clove relies on standard ECMP in physical switches, discovers paths using a novel traceroute mechanism, uses software-based flowlet-switching, and continuously learns congestion (or path utilization) state using standard switch features. It then manipulates packet-header fields in the hypervisor switch to direct traffic over less congested paths. Clove achieves 1.5 to 7 times smaller flow-completion times at 70% network load than other load-balancing algorithms that work with existing hardware. Clove also captures some 80% of the performance gain of best-of-breed hardware-based load-balancing algorithms like CONGA that require new equipment.

New congestion control algorithms are rapidly improving datacenters by reducing latency, overcoming incast, increasing throughput and improving fairness. Ideally, the operating system in every server and virtual machine is updated to support new congestion control algorithms. However, legacy applications often cannot be upgraded to a new operating system version, which means the advances are off-limits to them. Worse, as we show, legacy applications can be squeezed out, which in the worst case prevents the entire network from adopting new algorithms.

Yahoo mail servers have been receiving an enormous number of messages each day for the past 17 years. The vast majority of today's messages are machine-generated (about 90% of the messages), based on a boilerplate with a small number of specific per-recipient changes. We show that the popular Zlib compression to gzip format fails to fully utilize the high similarity between these machine-generated messages. In this paper we analyze the data redundancy in Yahoo mail, and present methods to reduce its space requirements while using the standard Zlib library. Our results show we can further reduce the compressed data size by a factor of almost 2.5, compared to traditional gzip compression.

Since mobile devices are equipped with limited battery power, it is essential to have efficient power management mechanisms in mobile broadband networks such as mobile WiMAX that enable always on connectivity and provide high throughputs. In recent times Voice over IP (VoIP) has emerged as an important and dominant application in broadband mobile networks such as WiMAX. VoIP traffic is uniquely characterized by small packets sent regularly at periodic intervals and are bound by strict latency requirements for satisfactory user experience. This paper presents novel mechanisms to achieve enhanced power saving during a VoIP session by exploiting unique characteristics of VoIP traffic in next generation mobile WiMAX networks based on IEEE 802.16m standard. The analytical model and simulation results indicate that proposed methods reduce the average power consumption during a VoIP call by up to 70% compared to the existing IEEE 802.16m draft specifications. Moreover, the simulation results closely match our proposed analytical model.