Towards Federated Learning using FaaS Fabric

Chadha, Mohak; Jindal, Anshul; Gerndt, Michael

doi:10.1145/3429880.3430100

Cited by 23 publications

(20 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The other benchmarks were not parallelized due to data and loop-carried dependencies in the implemented kernels. We use the tbb 2 Python package for TBB support.…”

Section: Optimizing and Maximizing Performance With Numbamentioning

confidence: 99%

“…Since the introduction of AWS Lambda [1] by Amazon in 2014, serverless computing has grown to support a wide variety of applications such as machine learning [2], map/reduce-style jobs [3], and compute-intensive scientific workloads [4], [5], [6], [7]. Function-as-a-Service (FaaS), a key enabler of serverless computing allows a traditional monolithic application to be decomposed into fine-grained functions that are executed in response to event triggers or HTTP requests [8] on a FaaS platform.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Architecture-Specific Performance Optimization of Compute-Intensive FaaS Functions

Chadha¹,

Jindal²,

Gerndt³

2021

Preprint

Self Cite

View full text Add to dashboard Cite

FaaS allows an application to be decomposed into functions that are executed on a FaaS platform. The FaaS platform is responsible for the resource provisioning of the functions. Recently, there is a growing trend towards the execution of compute-intensive FaaS functions that run for several seconds. However, due to the billing policies followed by commercial FaaS offerings, the execution of these functions can incur significantly higher costs. Moreover, due to the abstraction of underlying processor architectures on which the functions are executed, the optimization of these functions is challenging. As a result, most FaaS functions use pre-compiled libraries generic to x86-64 leading to performance degradation. In this paper, we examine the underlying processor architectures for Google Cloud Functions (GCF) and determine their prevalence across the 19 available GCF regions. We modify, adapt, and optimize a representative set of six compute-intensive FaaS workloads written in Python using Numba, a JIT compiler based on LLVM, and present results wrt performance, memory consumption, and costs on GCF. Results from our experiments show that the optimization of FaaS functions can improve performance by 18.2x (geometric mean) and save costs by 76.8% on average for the six functions. Our results show that optimization of the FaaS functions for the specific architecture is very important. We achieved a maximum speedup of 1.79x by tuning the function especially for the instruction set of the underlying processor architecture.

show abstract

“…The other benchmarks were not parallelized due to data and loop-carried dependencies in the implemented kernels. We use the tbb 2 Python package for TBB support.…”

Section: Optimizing and Maximizing Performance With Numbamentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Architecture-Specific Performance Optimization of Compute-Intensive FaaS Functions

Chadha¹,

Jindal²,

Gerndt³

2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…With the advent of Amazon Web Services (AWS) Lambda in 2014, serverless computing has gained popularity and more adoption in different application domains such as machine learning [1], [2], linear algebra computation [3], [4], and map/reduce-style jobs [5]. Furthermore, nowadays, it is implemented by every major cloud provider in services like Azure Functions [6] and Google Cloud Functions [7].…”

Section: Introductionmentioning

confidence: 99%

TppFaaS: Modeling Serverless Functions Invocations via Temporal Point Processes

et al. 2022

Self Cite

View full text Add to dashboard Cite

Serverless computing is a cloud computing paradigm that allows developers to focus exclusively on business logic as cloud service providers manage resource management tasks. Serverless applications based on this model are often composed of several fine-grained and ephemeral Function-as-a-Service (FaaS) functions that implement complex business processes via mutual interaction and interaction with Backend-as-a-Services (BaaS) such as databases. FaaS functions suffer from the cold start problem because of the scale to zero instances feature. In this work, we use neural Temporal Point Processes (TPPs) to model function invocations in FaaS compositions. A probability distribution over the time and class of the following invocations for a given history of invocations is predicted using these probabilistic models. The prediction can avoid cold starts by scaling functions in advance and reduce network load by optimizing the function-server assignment. In this regard, we developed a python-based tool called TppFaaS on top of OpenWhisk open-source serverless platform. TppFaaS uses the neural TPPs LogNormMix for modeling the time using a log-normal mixture distribution and TruncNorm for predicting a single value for the time. Furthermore, we built a custom trace data collector for OpenWhisk embedded into TppFaaS and created datasets for multiple FaaS compositions to train and test our models. For datasets without cold starts, the models achieved for most compositions a mean absolute error below 22ms and a percentage of correctly predicted function classes above 94%.

show abstract

“…In the context of serverless, FL clients are functions deployed on a FaaS platform that are capable of calculating the shared model updates. A connected group of different FaaS platforms is referred to as a FaaS fabric [12].…”

Section: Introductionmentioning

confidence: 99%

“…• We compare FedLess with an optimized reimplementation of an existing FaaS based FL system [12] and demonstrate that our system is significantly faster.…”

Section: Introductionmentioning

confidence: 99%

FedLess: Secure and Scalable Federated Learning Using Serverless Computing

Grafberger¹,

Chadha²,

Jindal³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

The traditional cloud-centric approach for Deep Learning (DL) requires training data to be collected and processed at a central server which is often challenging in privacysensitive domains like healthcare. Towards this, a new learning paradigm called Federated Learning (FL) has been proposed that brings the potential of DL to these domains while addressing privacy and data ownership issues. FL enables remote clients to learn a shared ML model while keeping the data local. However, conventional FL systems face several challenges such as scalability, complex infrastructure management, and wasted compute and incurred costs due to idle clients. These challenges of FL systems closely align with the core problems that serverless computing and Function-as-a-Service (FaaS) platforms aim to solve. These include rapid scalability, no infrastructure management, automatic scaling to zero for idle clients, and a payper-use billing model. To this end, we present a novel system and framework for serverless FL, called FedLess. Our system supports multiple commercial and self-hosted FaaS providers and can be deployed in the cloud, on-premise in institutional data centers, and on edge devices. To the best of our knowledge, we are the first to enable FL across a large fabric of heterogeneous FaaS providers while providing important features like security and Differential Privacy. We demonstrate with comprehensive experiments that the successful training of DNNs for different tasks across up to 200 client functions and more is easily possible using our system. Furthermore, we demonstrate the practical viability of our methodology by comparing it against a traditional FL system and show that it can be cheaper and more resourceefficient.

show abstract

Towards Federated Learning using FaaS Fabric

Cited by 23 publications

References 15 publications

Architecture-Specific Performance Optimization of Compute-Intensive FaaS Functions

Architecture-Specific Performance Optimization of Compute-Intensive FaaS Functions

TppFaaS: Modeling Serverless Functions Invocations via Temporal Point Processes

FedLess: Secure and Scalable Federated Learning Using Serverless Computing

Contact Info

Product

Resources

About