There is, I think, something ethereal about i —the square root of minus one. I remember first hearing about it at school. It seemed an odd beast at that time—an intruder hovering on the edge of reality.

Usually familiarity dulls this sense of the bizarre, but in the case of i it was the reverse: over the years the sense of its surreal nature intensified. It seemed that it was impossible to write mathematics that described the real world in …

I and i

Data Mining Practical Machine Learning Tools and Techniques

Social Network Analysis

The Grid 2: Blueprint for a New Computing Infrastructure

Picking the right cloud configuration for recurring big data analytics jobs running in clouds is hard, because there can be tens of possible VM instance types and even more cluster sizes to pick from. Choosing poorly can significantly degrade performance and increase the cost to run a job by 2-3x on average, and as much as 12x in the worst-case. However, it is challenging to automatically identify the best configuration for a broad spectrum of applications and cloud configurations with low search cost. CherryPick is a system that leverages Bayesian Optimization to build performance models for various applications, and the models are just accurate enough to distinguish the best or close-to-the-best configuration from the rest with only a few test runs. Our experiments on five analytic applications in AWS EC2 show that CherryPick has a 45-90% chance to find optimal configurations, otherwise near-optimal, saving up to 75% search cost compared to existing solutions.

/pdf/cherrypick-adaptively-unearthing-the-best-cloud-21u26pl50w.pdf

Cherrypick: adaptively unearthing the best cloud configurations for big data analytics

Web service providers have been using NoSQL datastores to provide scalability and availability for globally distributed data at the cost of sacrificing transactional guarantees. Recently, major web service providers like Google have moved towards building storage systems that provide ACID transactional guarantees for globally distributed data. For example, the newly published system, Spanner, uses Two-Phase Commit and Two-Phase Locking to provide atomicity and isolation for globally distributed data, running on top of Paxos to provide fault-tolerant log replication. We show in this paper that it is possible to provide the same ACID transactional guarantees for multi-datacenter databases with fewer cross-datacenter communication trips, compared to replicated logging. Instead of replicating the transactional log, we replicate the commit operation itself, by running Two-Phase Commit multiple times in different datacenters and using Paxos to reach consensus among datacenters as to whether the transaction should commit. Doing so not only replaces several inter-datacenter communication trips with intra-datacenter communication trips, but also allows us to integrate atomic commitment and isolation protocols with consistent replication protocols to further reduce the number of cross-datacenter communication trips needed for consistent replication; for example, by eliminating the need for an election phase in Paxos. We analyze our approach in terms of communication trips to compare it against the log replication approach, then we conduct an extensive experimental study to compare the performance and scalability of both approaches under various multi-datacenter setups.

http://www.nawab.me/Uploads/Replicated_Commit_VLDB_2013.pdf

Low-latency multi-datacenter databases using replicated commit

We present TritonSort, a highly efficient, scalable sorting system. It is designed to process large datasets, and has been evaluated against as much as 100 TB of input data spread across 832 disks in 52 nodes at a rate of 0.916 TB/min. When evaluated against the annual Indy GraySort sorting benchmark, TritonSort is 60% better in absolute performance and has over six times the per-node efficiency of the previous record holder. In this paper, we describe the hardware and software architecture necessary to operate TritonSort at this level of efficiency. Through careful management of system resources to ensure cross-resource balance, we are able to sort data at approximately 80% of the disks' aggregate sequential write speed.We believe the work holds a number of lessons for balanced system design and for scale-out architectures in general. While many interesting systems are able to scale linearly with additional servers, per-server performance can lag behind per-server capacity by more than an order of magnitude. Bridging the gap between high scalability and high performance would enable either significantly cheaper systems that are able to do the same work or provide the ability to address significantly larger problem sets with the same infrastructure.

/pdf/tritonsort-a-balanced-large-scale-sorting-system-1jif8yltdp.pdf

TritonSort: a balanced large-scale sorting system

A multitenant database management system (DBMS) in the cloud must continuously monitor the trade-off between efficient resource sharing among multiple application databases (tenants) and their performance. Considering the scale of \attn{hundreds to} thousands of tenants in such multitenant DBMSs, manual approaches for continuous monitoring are not tenable. A self-managing controller of a multitenant DBMS faces several challenges. For instance, how to characterize a tenant given its variety of workloads, how to reduce the impact of tenant colocation, and how to detect and mitigate a performance crisis where one or more tenants' desired service level objective (SLO) is not achieved.We present Delphi, a self-managing system controller for a multitenant DBMS, and Pythia, a technique to learn behavior through observation and supervision using DBMS-agnostic database level performance measures. Pythia accurately learns tenant behavior even when multiple tenants share a database process, learns good and bad tenant consolidation plans (or packings), and maintains a pertenant history to detect behavior changes. Delphi detects performance crises, and leverages Pythia to suggests remedial actions using a hill-climbing search algorithm to identify a new tenant placement strategy to mitigate violating SLOs. Our evaluation using a variety of tenant types and workloads shows that Pythia can learn a tenant's behavior with more than 92% accuracy and learn the quality of packings with more than 86% accuracy. During a performance crisis, Delphi is able to reduce 99th percentile latencies by 80%, and can consolidate 45% more tenants than a greedy baseline, which balances tenant load without modeling tenant behavior.

Characterizing tenant behavior for placement and crisis mitigation in multitenant DBMSs

We present TritonSort, a highly efficient, scalable sorting system. It is designed to process large datasets, and has been evaluated against as much as 100TB of input data spread across 832 disks in 52 nodes at a rate of 0.938TB/min. When evaluated against the annual Indy GraySort sorting benchmark, TritonSort is 66p better in absolute performance and has over six times the per-node throughput of the previous record holder. When evaluated against the 100TB Indy JouleSort benchmark, TritonSort sorted 9703 records/Joule. In this article, we describe the hardware and software architecture necessary to operate TritonSort at this level of efficiency. Through careful management of system resources to ensure cross-resource balance, we are able to sort data at approximately 80p of the disks’ aggregate sequential write speed.We believe the work holds a number of lessons for balanced system design and for scale-out architectures in general. While many interesting systems are able to scale linearly with additional servers, per-server performance can lag behind per-server capacity by more than an order of magnitude. Bridging the gap between high scalability and high performance would enable either significantly less expensive systems that are able to do the same work or provide the ability to address significantly larger problem sets with the same infrastructure.

/pdf/tritonsort-a-balanced-and-energy-efficient-large-scale-2eloo39nzx.pdf

TritonSort: A Balanced and Energy-Efficient Large-Scale Sorting System

As Android OS establishes itself as the primary platform on smartphones, a substantial increase in malware targeted at Android devices is being observed in the wild. While anti-virus software is available, and Android limits applications to user approved permissions, many users remain unaware of the threat posed by malware and of actual infections on their devices. In this paper we explore techniques to enable mobile network operators to detect Android malware and violations of user privacy through network traffic analysis.

Alexander Pucher

Papers

Low-latency multi-datacenter databases using replicated commit

TritonSort: a balanced large-scale sorting system

Characterizing tenant behavior for placement and crisis mitigation in multitenant DBMSs

TritonSort: A Balanced and Energy-Efficient Large-Scale Sorting System

Detecting Android Malware on Network Level