Fast deployment and execution of cloud functions in Function-as-a-Service (FaaS) platforms is critical, for example, microservices architectures. However, that require large packages or libraries are bloated start slowly. An optimization to cache at the worker nodes instead bundling them with functions. existing FaaS schedulers vanilla load balancers, agnostic cached response prior function executions, cannot properly reap benefits package caching. We study case package-aware scheduling...

We consider the problem of scheduling small cloud functions on serverless computing platforms. Fast deployment and execution these is critical, for example, microservices architectures. However, that require large packages or libraries are bloated start slowly. A solution to cache at worker nodes instead bundling them with functions. existing FaaS schedulers vanilla load balancers, agnostic any may have been cached in response prior function executions, cannot reap benefits package caching...

Cloud computing providers offer a variety of pricing models, complicating the client decision, as no single model is cheapest in all scenarios. In addition, small to medium-sized organizations frequently lack personnel that can navigate intricacies each model, and result, end up opting for sub-optimal strategy, leading overpaying resources or not being able meet performance goals. this paper, we: (1) present results study shows that, Ecuador, considerable percentage companies choose...

We study the workload of an Online Invoicing application with clients in Andean region South America. The is offered to a Software-as-a-Service model, has microservices architecture and runs on containerized environment public cloud provider. described this paper can be used as part suite comprised different workloads, when evaluating architectures. To best our knowledge, novel web domain it complements other workloads publicly available. Though we make no claim general applicability...

Cloud-native applications are increasingly adopting microservices architectures that support the development agility required by modern software. These deploy their components in containers enable to be deployed across different platforms, supporting independent scaling of components. However, operational complexity presents significant challenges maintaining performance such applications, especially clouds with variability and unpredictability. While virtual machine based deployment has...

Key-value stores are an important component of cloud applications. These NoSQL databases typically do not support ACID transactions, hence, their applications have diverged from traditional database workloads and well represented by benchmarks like TPC-C. In this context, YCSB emerged as the de facto benchmark for serving stores, supporting a wide variety synthetic workloads. However, no publicly available tool exists that can replay real traces against these systems, is harmful to...

While significant advances have been made towards realizing self-tuning cloud caches, existing products still require manual tuning. These systems are built to serve requests extremely fast and anything that consumes resources not directly related the request-serving control path is avoided. We show severless computing platforms can be leveraged solve complex optimization problems arise during loops, thus used optimize in for free. To our approach feasible useful, we implement SPREDS...

The research community has made significant advances towards realizing self-tuning cloud caches; notwithstanding, existing products still require manual expert tuning to maximize performance. Cloud (software) caches are built swiftly serve requests; thus, avoiding costly functionality additions not directly related the request-serving control path is critical. We show that serverless computing services can be leveraged solve complex optimization problems arise during loops and used optimize...

Correctly applying distributed systems concepts is important for software that seeks to be scalable, reliable and fast. For this reason, Distributed Systems a course included in many Computer Science programs. To both describe current trends teaching as reference educators seek improve the quality of their syllabi, we present review 51 syllabi courses from top programs around world. We manually curated extracted data allowed us identify approaches used subject, including choice topics, book,...

We study the performance of cache admission and eviction policies under workloads with malicious requests (i.e., that try to poison or pollute useless information). The we have chosen for this are: LRU, SLRU, SLRU2. also evaluated W-TinyLFU, a frequency-based policy seeks increase effectiveness cache, specially skewed access distributions. implemented in caching simulator them using Zipfian real from Yahoo, YouTube, feature animation company. In addition, created modified versions these...

We analyze a dataset of 51 current (2019-2020) Distributed Systems syllabi from top Computer Science programs, focusing on finding the prevalence and context in which topics related to performance are being taught these courses. also study scale infrastructure mentioned DS courses, small client-server systems cloud-scale, peer-to-peer, global-scale systems. make eight main findings, covering goals such as performance, scalability its variant elasticity; activities benchmarking monitoring;...

Cloud computing services can significantly reduce an organization's IT cost while improving the fault tolerance and scalability of its services. However, developing emerging markets have been slow in adopting cloud technologies. We report on results a study state adoption Ecuador. analyze identify challenges opportunities that arise from current adoption. Our be leveraged by providers to design better strategies increase their participation markets, as well companies market solutions.