Microservice-Based Workflows (MBWs) are used popularly to govern the composition and coordination of individual microservices to realize business processes. With MBWs, designers often aim to maximize concurrency to increase the chances of successful workflow collaboration and enhance business process efficiency. However, operations involving manipulating and accessing artifacts (data objects) within these workflows may introduce anomalies leading to unexpected artifact states. In workflow design phase, seeking bug-free artifact states is vital to help prevent crashes, errors, and unexpected outcomes during execution. Concurrent artifact anomalies are referred to abnormal parallel operations on the same artifact. Few studies have explored the detection of concurrent anomalies in MBWs, and they are inefficient and ineffective as they struggle with high time complexity and are insensitive to the presence of nested AND gateways. This paper focuses on improving microservices workflows reliability by detecting concurrent anomalies in artifacts during the design. We present a series of methods to detect the anomalies based on SP-tree, a tree structure to record workflow paths and artifact information. Our methods outperform existing ones by detecting more anomalies with lower time and space complexity.

During the design phase of a workflow process, detecting anomalous operations on artifacts is important for preventing errors and unexpected behaviors of the process dynamically. For temporal structured workflows (TSWs), which involve specifying the min and max execution time intervals for each activity, there are few studies on analyzing artifact anomalies compared to non-temporal workflows. Additionally, the existing approaches designed for TSWs are inaccurate in detecting these anomalies either. To improve the analysis of TSWs, this paper presents an improved methodology based on an extended SP-tree structure, called TSP-tree. Our approach involves two steps: first, transforming a TSW into a TSP-tree, and then applying several algorithms to TSP-tree to detect artifact anomalies. Compared to previous methods, our approach provides lower time complexity for detection, simplifies the analysis, and detects more artifact anomalies

With the advent of the Service Oriented Architecture (SOA) in system design, various domain knowledges are included in a service-based application, such as the design of Artificial Intelligence (AI) or augmented reality (AR) systems. While merging one or multiple domains into computation systems, the computation systems can be widely applied in various domain usages with novelty, useful, and surprising properties, which are defined as systems of creative computing. In creative computing, several theoretical evaluation metrics and verification approaches have been proposed for system design in several domains. However, a solid practical design environment for creative service-based systems is rarely considered in current researches. In this paper, we propose a model for creative service software development based on semantic web, which is applied in two phases: (1) requirement specification and (2) service design. In order to bridge the knowledge gap between domain experts and software engineers, and provide a machine-readable format for creative computing, two sub-models, Requirement Specification and Service Structure Models, are constructed in both phases, sequentially. After the latter sub-model is validated, the creative service software is well-constructed based on the services definition and composition represented by the model.

Workflow management systems (WfMS) are considered as accomplish platforms which can provide structured organization in business process and service architecture design. The systems contain workflow models in foundation, which provide flow control in one or more task sequences in parallel. Manipulation and access of artifacts that occur in or between the task sequences can generate unexpected state of artifacts by inappropriate workflow design. The artifact anomalies in a workflow model are classified into two categories, which are types of continuous and concurrent anomalies. A continuous anomaly occurs while an artifact is written redundantly or accessed before production. On the other hand, a concurrent anomaly can occur while an artifact is conflict written in parallel in a workflow model. There are several methods presented for anomaly analysis, however, these methods cannot detect all anomalies due to their definitions and they are either inefficient or lack of proof for the correctness. In this article, we present improved detection methods with an improved C-tree structure, called SP-tree. Based on an updated anomaly definition, our anomaly detection includes two stages: 1) the transformation algorithm generates an equivalent SP-tree from a given structured workflow model; and 2) based on the generated equivalent SP-tree, a series of methods are applied to detect anomalies.

The microservice architecture is a Service-Oriented Architecture (SOA) where a service-based application can be composed of a number of smaller but independently concurrent running units, called microservices, to improve the performance and maintainability of the application. In an application with microservice architecture, an unexpected artifact state(s) inside a microservice may be exchanged to another microservice or other service units and corrupt the whole system of the application. On the other hand, the abnormal artifact operation pairs can be categorized into continuous and concurrent artifact anomalies which indicate that two sequential and parallel operations working on the same artifact resulting in abnormal behavior semantically. The recent studies showed that an SP-tree structure adopted in the detections of both anomalies inside a structured workflow can reduce the computation complexity of detection as linear. In this paper, we present a series of methods based on SP-tree to detect the artifact anomalies inside each microservice of the application during the design phase. Different from the design of applications with traditional services, where each service is assumed to contain all possible types of anomalies and cannot be modified directly, the designer of microservice is concerned with the limited scope and can modify each microservice based on the anomaly detection results. Our contribution includes identifying the artifact properties in a microservice architecture and the methods to detect the anomalies based on these properties which can simplify the detection of artifact anomaly in service-based applications.

Workflow models are widely applied in business software design. A workflow model contains a set of systematic ordered tasks to achieve designated business goal(s) under the designed flow control. Analyzing artifact usage during design phase can prevent unexpected artifact result due to abnormal artifact operation(s). A sequential anomaly indicates a pair of activities operating on the same artifact that can result in redundant write or missing production. On the other hand, the iteration of a loop structure in a workflow cannot be statically analyzed, thus, detecting process of artifact anomalies in a loop is costly. In this paper, we present an effective method to detect all anomalies associated with a loop by removing the redundant computation due to the repeated structure of the body and control in the iterations. After the removing, the anomalies can be detected on a single iteration generated instead. Here, the process of anomaly detection is now simplified into two phases: First, a workflow model is transformed into a corresponding C-tree structure and next, the proposed anomaly detection methodology is applied to the C-tree. Compared with current approaches, our method can reduce the space complexity and decrease the execution times of anomaly detection as linear.

Previous studies have used on-chip thermal sensors (diodes) to estimate the leakage power of a CPU. However, an embedded CPU equips only a few thermal sensors and may suffer from considerable spatial temperature variances across the CPU core, and leakage power estimation based on insufficient temperature information introduces errors. According to our experiments, the conventional leakage power models may have up to 22.9% estimation error for a 70-nm embedded CPU. In this study, we first evaluated the accuracy of leakage power estimates based on thermal sensors on different locations of a CPU and suggested locations that can reduce the error to 0.9%. Then, we proposed temperature-referred and counter-tracked estimation (TRACE) that relies on temperature sensors and hardware activity counters to estimate leakage power. The simulation results demonstrated that employing TRACE could reduce the error to 3.4%. Experiments were also conducted on a real platform to verify our findings.

In recent years, thermal management, which improves the reliability, performance, power leakage, etc. of modern microprocessors, has been the subject of numerous computer architecture and system software studies. To determine the detailed thermal distribution of a microprocessor is among the critical tasks for thermal management. However, because thermal modeling tools require considerable computation time and memory to simulate fine-grain thermal information, they may be unsuitable for dynamic thermal management and hardware implementation. This study proposes a novel model based on reduced resistance-capacitance (RC) networks for efficiently calculating the temperature of a microprocessor. The proposed model is compared with two existing thermal simulation tools, namely, HotSpot [1] and Temptor [2]. The experiment studies show that the results generated using the proposed model differ from those of the existing tools by only 0.5 to 1.5%. However, the suggested model can increase computation speeds by 5 to 9 times and 98 to 161 times that of Temptor and HotSpot, respectively. For the memory usage, the proposed model consumes merely 0.45% of the space used by the existing tools.