-
Informed and Assessable Observability Design Decisions in Cloud-native Microservice Applications
Authors:
Maria C. Borges,
Joshua Bauer,
Sebastian Werner,
Michael Gebauer,
Stefan Tai
Abstract:
Observability is important to ensure the reliability of microservice applications. These applications are often prone to failures, since they have many independent services deployed on heterogeneous environments. When employed "correctly", observability can help developers identify and troubleshoot faults quickly. However, instrumenting and configuring the observability of a microservice applicati…
▽ More
Observability is important to ensure the reliability of microservice applications. These applications are often prone to failures, since they have many independent services deployed on heterogeneous environments. When employed "correctly", observability can help developers identify and troubleshoot faults quickly. However, instrumenting and configuring the observability of a microservice application is not trivial but tool-dependent and tied to costs. Architects need to understand observability-related trade-offs in order to weigh between different observability design alternatives. Still, these architectural design decisions are not supported by systematic methods and typically just rely on "professional intuition". In this paper, we argue for a systematic method to arrive at informed and continuously assessable observability design decisions. Specifically, we focus on fault observability of cloud-native microservice applications, and turn this into a testable and quantifiable property. Towards our goal, we first model the scale and scope of observability design decisions across the cloud-native stack. Then, we propose observability metrics which can be determined for any microservice application through so-called observability experiments. We present a proof-of-concept implementation of our experiment tool OXN. OXN is able to inject arbitrary faults into an application, similar to Chaos Engineering, but also possesses the unique capability to modify the observability configuration, allowing for the assessment of design decisions that were previously left unexplored. We demonstrate our approach using a popular open source microservice application and show the trade-offs involved in different observability design decisions.
△ Less
Submitted 1 March, 2024;
originally announced March 2024.
-
Selective Prediction for Semantic Segmentation using Post-Hoc Confidence Estimation and Its Performance under Distribution Shift
Authors:
Bruno Laboissiere Camargos Borges,
Bruno Machado Pacheco,
Danilo Silva
Abstract:
Semantic segmentation plays a crucial role in various computer vision applications, yet its efficacy is often hindered by the lack of high-quality labeled data. To address this challenge, a common strategy is to leverage models trained on data from different populations, such as publicly available datasets. This approach, however, leads to the distribution shift problem, presenting a reduced perfo…
▽ More
Semantic segmentation plays a crucial role in various computer vision applications, yet its efficacy is often hindered by the lack of high-quality labeled data. To address this challenge, a common strategy is to leverage models trained on data from different populations, such as publicly available datasets. This approach, however, leads to the distribution shift problem, presenting a reduced performance on the population of interest. In scenarios where model errors can have significant consequences, selective prediction methods offer a means to mitigate risks and reduce reliance on expert supervision. This paper investigates selective prediction for semantic segmentation in low-resource settings, thus focusing on post-hoc confidence estimators applied to pre-trained models operating under distribution shift. We propose a novel image-level confidence measure tailored for semantic segmentation and demonstrate its effectiveness through experiments on three medical imaging tasks. Our findings show that post-hoc confidence estimators offer a cost-effective approach to reducing the impacts of distribution shift.
△ Less
Submitted 6 May, 2024; v1 submitted 16 February, 2024;
originally announced February 2024.
-
DeepLearningBrasil@LT-EDI-2023: Exploring Deep Learning Techniques for Detecting Depression in Social Media Text
Authors:
Eduardo Garcia,
Juliana Gomes,
Adalberto Barbosa Júnior,
Cardeque Borges,
Nádia da Silva
Abstract:
In this paper, we delineate the strategy employed by our team, DeepLearningBrasil, which secured us the first place in the shared task DepSign-LT-EDI@RANLP-2023, achieving a 47.0% Macro F1-Score and a notable 2.4% advantage. The task was to classify social media texts into three distinct levels of depression - "not depressed," "moderately depressed," and "severely depressed." Leveraging the power…
▽ More
In this paper, we delineate the strategy employed by our team, DeepLearningBrasil, which secured us the first place in the shared task DepSign-LT-EDI@RANLP-2023, achieving a 47.0% Macro F1-Score and a notable 2.4% advantage. The task was to classify social media texts into three distinct levels of depression - "not depressed," "moderately depressed," and "severely depressed." Leveraging the power of the RoBERTa and DeBERTa models, we further pre-trained them on a collected Reddit dataset, specifically curated from mental health-related Reddit's communities (Subreddits), leading to an enhanced understanding of nuanced mental health discourse. To address lengthy textual data, we used truncation techniques that retained the essence of the content by focusing on its beginnings and endings. Our model was robust against unbalanced data by incorporating sample weights into the loss. Cross-validation and ensemble techniques were then employed to combine our k-fold trained models, delivering an optimal solution. The accompanying code is made available for transparency and further development.
△ Less
Submitted 8 November, 2023;
originally announced November 2023.
-
A Neural Network Warm-Start Approach for the Inverse Acoustic Obstacle Scattering Problem
Authors:
Mo Zhou,
Jiequn Han,
Manas Rachh,
Carlos Borges
Abstract:
We consider the inverse acoustic obstacle problem for sound-soft star-shaped obstacles in two dimensions wherein the boundary of the obstacle is determined from measurements of the scattered field at a collection of receivers outside the object. One of the standard approaches for solving this problem is to reformulate it as an optimization problem: finding the boundary of the domain that minimizes…
▽ More
We consider the inverse acoustic obstacle problem for sound-soft star-shaped obstacles in two dimensions wherein the boundary of the obstacle is determined from measurements of the scattered field at a collection of receivers outside the object. One of the standard approaches for solving this problem is to reformulate it as an optimization problem: finding the boundary of the domain that minimizes the $L^2$ distance between computed values of the scattered field and the given measurement data. The optimization problem is computationally challenging since the local set of convexity shrinks with increasing frequency and results in an increasing number of local minima in the vicinity of the true solution. In many practical experimental settings, low frequency measurements are unavailable due to limitations of the experimental setup or the sensors used for measurement. Thus, obtaining a good initial guess for the optimization problem plays a vital role in this environment.
We present a neural network warm-start approach for solving the inverse scattering problem, where an initial guess for the optimization problem is obtained using a trained neural network. We demonstrate the effectiveness of our method with several numerical examples. For high frequency problems, this approach outperforms traditional iterative methods such as Gauss-Newton initialized without any prior (i.e., initialized using a unit circle), or initialized using the solution of a direct method such as the linear sampling method. The algorithm remains robust to noise in the scattered field measurements and also converges to the true solution for limited aperture data. However, the number of training samples required to train the neural network scales exponentially in frequency and the complexity of the obstacles considered. We conclude with a discussion of this phenomenon and potential directions for future research.
△ Less
Submitted 3 August, 2023; v1 submitted 16 December, 2022;
originally announced December 2022.
-
FaaSter Troubleshooting -- Evaluating Distributed Tracing Approaches for Serverless Applications
Authors:
Maria C. Borges,
Sebastian Werner,
Ahmet Kilic
Abstract:
Serverless applications can be particularly difficult to troubleshoot, as these applications are often composed of various managed and partly managed services. Faults are often unpredictable and can occur at multiple points, even in simple compositions. Each additional function or service in a serverless composition introduces a new possible fault source and a new layer to obfuscate faults. Curren…
▽ More
Serverless applications can be particularly difficult to troubleshoot, as these applications are often composed of various managed and partly managed services. Faults are often unpredictable and can occur at multiple points, even in simple compositions. Each additional function or service in a serverless composition introduces a new possible fault source and a new layer to obfuscate faults. Currently, serverless platforms offer only limited support for identifying runtime faults. Developers looking to observe their serverless compositions often have to rely on scattered logs and ambiguous error messages to pinpoint root causes. In this paper, we investigate the use of distributed tracing for improving the observability of faults in serverless applications. To this end, we first introduce a model for characterizing fault observability, then provide a prototypical tracing implementation - specifically, a developer-driven and a platform-supported tracing approach. We compare both approaches with our model, measure associated trade-offs (execution latency, resource utilization), and contribute new insights for troubleshooting serverless compositions.
△ Less
Submitted 7 October, 2021;
originally announced October 2021.
-
TIRA: An OpenAPI Extension and Toolbox for GDPR Transparency in RESTful Architectures
Authors:
Elias Grünewald,
Paul Wille,
Frank Pallas,
Maria C. Borges,
Max-R. Ulbricht
Abstract:
Transparency - the provision of information about what personal data is collected for which purposes, how long it is stored, or to which parties it is transferred - is one of the core privacy principles underlying regulations such as the GDPR. Technical approaches for implementing transparency in practice are, however, only rarely considered. In this paper, we present a novel approach for doing so…
▽ More
Transparency - the provision of information about what personal data is collected for which purposes, how long it is stored, or to which parties it is transferred - is one of the core privacy principles underlying regulations such as the GDPR. Technical approaches for implementing transparency in practice are, however, only rarely considered. In this paper, we present a novel approach for doing so in current, RESTful application architectures and in line with prevailing agile and DevOps-driven practices. For this purpose, we introduce 1) a transparency-focused extension of OpenAPI specifications that allows individual service descriptions to be enriched with transparency-related annotations in a bottom-up fashion and 2) a set of higher-order tools for aggregating respective information across multiple, interdependent services and for coherently integrating our approach into automated CI/CD-pipelines. Together, these building blocks pave the way for providing transparency information that is more specific and at the same time better reflects the actual implementation givens within complex service architectures than current, overly broad privacy statements.
△ Less
Submitted 10 June, 2021;
originally announced June 2021.
-
A step toward a reinforcement learning de novo genome assembler
Authors:
Kleber Padovani,
Roberto Xavier,
Rafael Cabral Borges,
Andre Carvalho,
Anna Reali,
Annie Chateau,
Ronnie Alves
Abstract:
De novo genome assembly is a relevant but computationally complex task in genomics. Although de novo assemblers have been used successfully in several genomics projects, there is still no 'best assembler', and the choice and setup of assemblers still rely on bioinformatics experts. Thus, as with other computationally complex problems, machine learning may emerge as an alternative (or complementary…
▽ More
De novo genome assembly is a relevant but computationally complex task in genomics. Although de novo assemblers have been used successfully in several genomics projects, there is still no 'best assembler', and the choice and setup of assemblers still rely on bioinformatics experts. Thus, as with other computationally complex problems, machine learning may emerge as an alternative (or complementary) way for develo** more accurate and automated assemblers. Reinforcement learning has proven promising for solving complex activities without supervision - such games - and there is a pressing need to understand the limits of this approach to 'real' problems, such as the DFA problem. This study aimed to shed light on the application of machine learning, using reinforcement learning (RL), in genome assembly. We expanded upon the sole previous approach found in the literature to solve this problem by carefully exploring the learning aspects of the proposed intelligent agent, which uses the Q-learning algorithm, and we provided insights for the next steps of automated genome assembly development. We improved the reward system and optimized the exploration of the state space based on pruning and in collaboration with evolutionary computing. We tested the new approaches on 23 new larger environments, which are all available on the internet. Our results suggest consistent performance progress; however, we also found limitations, especially concerning the high dimensionality of state and action spaces. Finally, we discuss paths for achieving efficient and automated genome assembly in real scenarios considering successful RL applications - including deep reinforcement learning.
△ Less
Submitted 7 March, 2024; v1 submitted 2 February, 2021;
originally announced February 2021.
-
New methodology for facilitating the food wastage quantification. Identifying gaps and data inconsistencies
Authors:
Hector Barco Cobalea,
Iraia Oribe Garcia,
Maria Virginia Vargas Viedma,
Cruz Enrique Borges,
Cristina Martin Andonegui,
Ainhoa Alonso Vicario
Abstract:
The work aims at providing a new methodology to facilitate the process of quantifying the food waste according to European standards all along the agrifood chain combining information that is becoming available at local level.
This new methodology generates straightforward and easy-to-interpret results for the decision making process in the framework of the quantification of the food waste at lo…
▽ More
The work aims at providing a new methodology to facilitate the process of quantifying the food waste according to European standards all along the agrifood chain combining information that is becoming available at local level.
This new methodology generates straightforward and easy-to-interpret results for the decision making process in the framework of the quantification of the food waste at local and supralocal scale and it provides adequate procedures which are easy adaptable to the specific circumstances in each municipality. Moreover, this method could have applications for larger territorial contexts, as the national scale, detecting possible points for improvement of the current official figures at this respect.
△ Less
Submitted 6 May, 2019; v1 submitted 7 March, 2019;
originally announced March 2019.
-
RALL - Routing-Aware Of Path Length, Link Quality, And Traffic Load For Wireless Sensor Networks
Authors:
Vinicius N. Medeiros,
Douglas V. Santana,
Bruno Silvestre,
Vinicius da C. M. Borges
Abstract:
Due to the enormous variety of application scenarios and ubiquity,Internet of Things (IoT) brought a new perspective of applications for the current and future Internet. The Wireless Sensor Networks provide key devices for develo** the IoT communication paradigm, such as the sensors collecting various kind of information and the routing and MAC protocols. However, this type of network has strong…
▽ More
Due to the enormous variety of application scenarios and ubiquity,Internet of Things (IoT) brought a new perspective of applications for the current and future Internet. The Wireless Sensor Networks provide key devices for develo** the IoT communication paradigm, such as the sensors collecting various kind of information and the routing and MAC protocols. However, this type of network has strong power consumption and transmission capacity restrictions (low speed wireless links and subject to interference). In this context, it is necessary to develop solutions that enable a more efficient communication based on the optimized utilization of the network resources. This papers aims to present a multi-objective routing algorithm, named Routing-Aware of path Length, Link quality, and traffic Load (RALL), that seeks to balance three objectives: to minimize bottlenecks, to minimize path length, and to avoid links with low quality. RALL results in good performance when taking into consideration delivery rate, overhead, delay, and power consumption.
△ Less
Submitted 15 September, 2016;
originally announced September 2016.
