Runtime System Research Articles

Tug-of-War in the Selection of Materials for Battery Technologies

Batteries are the heart and the bottleneck of portable electronic systems. They power electronics and determine the system run time, with the size and volume determining factors in their design and implementation. Understanding the material properties of the battery components—anode, cathode, electrolyte, and separator—and their interaction is necessary to establish selection criteria based on their correlations with the battery metrics: capacity, current density, and cycle life. This review studies material used in the four battery components from the perspective and the impact of seven ions (Li+, Na+, K+, Zn2+, Ca2+, Mg2+, and Al3+), employed in commercial and research batteries. In addition, critical factors of sustainability of the supply chains—geographical raw materials origins vs. battery manufacturing companies and material properties (Young’s modulus vs. electric conductivity)—are mapped. These are key aspects toward identifying the supply chain vulnerabilities and gaps for batteries. In addition, two battery applications, smartphones and electric vehicles, in light of challenges in the current research, commercial fronts, and technical prospects, are discussed. Bringing the next generation of batteries necessitates a transition from advances in material to addressing the technical challenges, which the review has powered.

Adaptive Resource Efficient Microservice Deployment in Cloud-Edge Continuum

User-facing services are now evolving towards the microservice architecture where a service is built by connecting multiple microservice stages. Since the entire service is heavy, the microservice architecture shows the opportunity to only offload some microservice stages to the edge devices that are close to the end users. However, emerging techniques often result in the violation of Quality-of-Service (QoS) of microservice-based services in cloud-edge continuum, as they do not consider the communication overhead or the resource contention between microservices and external co-located tasks. We propose Nautilus, a runtime system that effectively deploys microservice-based user-facing services in cloud-edge continuum. Nautilus ensures the QoS of microservice-based user-facing services while minimizing the required computational resources, which is comprised of a communication-aware microservice mapper, a contention-aware resource manager and an IO-sensitive and load-aware microservice migration scheduler. The mapper divides the microservice graph into multiple partitions based on the communication overhead and maps the partitions to appropriate nodes. On each node, the resource manager determines the optimal resource allocation for its microservices based on reinforcement learning that may capture the complex contention behaviors. Once the microservices are suffered from external IO pressure, the IO-sensitive microservice scheduler migrates the critical one to idle nodes. Furthermore, when the load of microservices changes dynamically, the load-aware microservice scheduler migrates microservices from busy nodes to idle ones to ensure the QoS goal of the entire service. Our experimental results show that Nautilus can guarantee the required QoS target under external shared resources contention while the state-of-the-art suffers from QoS violations. Meanwhile, Nautilus reduces the computational resource usage by 23.9% and the network bandwidth usage by 53.4%, while achieving the required 99%-ile latency.

Mashing load balancing algorithm to boost hybrid kernels in molecular dynamics simulations

The path to the efficient exploitation of molecular dynamics simulators is strongly driven by the increasingly intensive use of accelerators. However, they suffer performance portability issues, making it necessary both to achieve technological combinations that allow taking advantage of each programming model and device, and to define more effective load distribution strategies that consider the simulation conditions. In this work, a new load balancing algorithm is presented, together with a set of optimizations to support hybrid co-execution in a runtime system for heterogeneous computing. The new extended design enables the exploitation of custom kernels and acceleration technologies altogether, being encapsulated for the rest of the runtime and its scheduling system. With this support, Mash algorithm allows to simultaneously leverage different workload distribution strategies, benefiting from the most advantageous one per device and technology. Experiments show that these proposals achieve an efficiency close to 0.90 and an energy efficiency improvement around 1.80 over the original optimized version.

Design and Analysis of a Smart Sensor-Based Early Warning Intervention Network for School Sports Bullying among Left-Behind Children

This paper constructs a sports bullying early warning intervention system using smart sensors to conduct in-depth research and analysis on early warning intervention of school sports bullying behaviors among left-behind children. Unlike daily behavior recognition based on motion sensors, school sports bullying actions are very random and difficult to be described by a specific motion trajectory. For the characteristics of violent actions and daily actions, action features in the time and frequency domains are extracted and action categories are recognized by BP neural networks; for complex actions, it is proposed to decompose complex actions into basic actions to improve the recognition rate. The algorithm of combining action features and speech features to achieve violence recognition is proposed. For the complexity of audio data features, this paper firstly preprocesses the audio data with preweighting, framing, and windowing and secondly extracts the MFCC feature parameters from the audio data and then builds a deep convolutional neural network to design the violence emotion recognition algorithm. The simulation results show that the algorithm effectively improves the accuracy rate of violent action recognition to 91.25% and the recall rate of violent action recognition to 92.13%. Finally, the LDA dimensionality reduction algorithm is introduced to address the problem of the high complexity of the algorithm due to the high number of feature dimensions. The LDA dimensionality reduction algorithm reduces the number of feature dimensions to 7 dimensions, which reduces the system running time by about 52% and improves the recognition rate of specific complex actions by about 12.1% while ensuring the overall system performance.

Using Barrier Elision to Improve Transactional Code Generation

With chip manufacturers such as Intel, IBM, and ARM offering native support for transactional memory in their instruction set architectures, memory transactions are on the verge of being considered a genuine application tool rather than just an interesting research topic. Despite this recent increase in popularity on the hardware side of transactional memory (HTM) , software support for transactional memory (STM) is still scarce and the only compiler with transactional support currently available, the GNU Compiler Collection (GCC) , does not generate code that achieves desirable performance. For hybrid solutions of TM (HyTM) , which are frameworks that leverage the best aspects of HTM and STM, the subpar performance of the software side, caused by inefficient compiler generated code, might forbid HyTM to offer optimal results. This article extends previous work focused exclusively on STM implementations by presenting a detailed analysis of transactional code generated by GCC in the context of HybridTM implementations. In particular, it builds on previous research of transactional memory support in the Clang/LLVM compiler framework, which is decoupled from any TM runtime, and presents the following novel contributions: (a) it shows that STM’s performance overhead, due to an excessive amount of read and write barriers added by the compiler, also impacts the performance of HyTM systems; and (b) it reveals the importance of the previously proposed annotation mechanism to reduce the performance gap between HTM and STM in phased runtime systems. Furthermore, it shows that, by correctly using the annotations on just a few lines of code, it is possible to reduce the total number of instrumented barriers by 95% and to achieve speed-ups of up to 7× when compared to the original code generated by GCC and the Clang compiler. 1

Online product sentiment analysis using random evolutionary whale optimization algorithm and deep belief network

• Image and text normalization techniques are used to improve the data quality. • Used MLBP , SURF, TF-IDF and LSA techniques for feature extraction. • Used RE-WOA to reduce the dimension of the extracted feature vectors. • Used DBN classifier to classify the user's sentiments of the online products. In the recent decades, the online product sentiment analysis is an emerging research topic that assists the customers to take better decisions on purchasing the products and to achieve better sales of the products. Recently, several machine learning techniques are experimented on many datasets for analyzing the customer's sentiments through online portals. Still, the customers are struggling to obtain the aspect sentiments expressed by other customers, particularly in the amazon websites. Therefore, a novel automated model is proposed in this manuscript for an effective online product sentiment analysis. After collecting the multimodal data from the amazon websites, the image and data normalization techniques are employed for better understanding of the collected data. Further, the feature extraction is performed by utilizing Latent Semantic Analysis (LSA), Term Frequency- Inverse Document Frequency (TF-IDF), Modified Local Binary Pattern (MLBP), and Speeded Up Robust Features (SURF) descriptors for extracting the textual and visual feature vectors from the preprocessed data. Finally, the Random Evolutionary Whale Optimization Algorithm (REWOA) and Deep Belief Network (DBN) classifier are integrated for feature vector optimization and sentiment classification. By using feature optimization, the system complexity and running time of the classifier is improved. The experimental investigation states that the developed REWOA-DBN model achieved 96.86% of classification accuracy, which is better compared to other optimizers and classifiers

Artificial Intelligence Driven Chatbot

Today, most large-scale conversational AI agents (e.g. Alexa, Siri, or Google Assistant) are built using manually annotated data to train the different components of the system. Typically, the accuracy of the ML models in these components are improved by manually transcribing and annotating data. As the scope of these systems increase to cover more scenarios and domains, manual annotation to improve the accuracy of these components becomes prohibitively costly and time- consuming. In this paper, a group of Amazon researchers propose a system that leverages user-system interaction feedback signals to automate learning without any manual annotation. Users here tend to modify a previous query in hopes of fixing an error in the previous turn to get the right results. These reformulations, which are often preceded by defective experiences caused by errors in ASR, NLU, ER or the application. In some cases, users may not properly formulate their requests (e.g. providing partial title of a song), but gleaning across a wider pool of users and sessions reveals the underlying recurrent patterns. The proposed self-learning system automatically detects the errors, generate reformulations and deploys fixes to the runtime system to correct different types of errors occurring in different components of the system. The results show that the approach is highly scalable, and able to learn reformulations that reduce Alexa-user errors by pooling anonymized data across millions of customers.

The Construction of an Action-Speech Feature-Based School Violence Recognition Algorithm and Occupational Therapy Education Model for Adolescents.

This paper constructs an algorithm for youth school violence recognition and an occupational therapy education model for victims through the extraction of action speech features. For the characteristics of violent actions and daily actions, action features in time and frequency domains are extracted and action categories are recognized by BP neural network; for complex actions, it is proposed to decompose complex actions into basic actions to improve the recognition rate; then, LDA dimensionality reduction algorithm is introduced for the problem of the high complexity of algorithm due to high dimensionality of features, and the feature dimensionality is reduced to 8 dimensions by LDA dimensionality reduction algorithm, which reduces the system running time by about 51% and improves the accuracy of violent action recognition by 3.3% while ensuring the overall performance of the system. The LDA dimensionality reduction algorithm reduces the number of features to 8 dimensions, which reduces the running time of the system by 51%, increases the accuracy rate of violent action recognition by 3.3%, and increases the recall rate of violent action recognition by 8.86% while ensuring the overall performance of the system. Based on the classical D-S theory, we proposed an improved D-S evidence fusion algorithm by modifying the original evidence model with a new probability distribution function and constructing new fusion rules, which can solve the fusion conflict problem well. The recall rate for violent actions is increased to 90.0%, thus reducing the missed alarm rate of the system.

Resource allocation for task-level speculative scientific applications: A proof of concept using Parallel Trajectory Splicing

The constant increase in parallelism available on large-scale distributed computers poses major scalability challenges to many scientific applications. A common strategy to improve scalability is to express algorithms in terms of independent tasks that can be executed concurrently on a runtime system. In this manuscript, we consider a generalization of this approach where task-level speculation is allowed. In this context, a probability is attached to each task which corresponds to the likelihood that the output of the speculative task will be consumed as part of the larger calculation. We consider the problem of optimal resource allocation to each of the possible tasks so as to maximize the total expected computational throughput. The power of this approach is demonstrated by analyzing its application to Parallel Trajectory Splicing, a massively-parallel long-time-dynamics method for atomistic simulations.

A scalability study of the Ice-sheet and Sea-level System Model (ISSM, version 4.18)

Abstract. Accurately modelling the contribution of Greenland and Antarctica to sea level rise requires solving partial differential equations at a high spatial resolution. In this paper, we discuss the scaling of the Ice-sheet and Sea-level System Model (ISSM) applied to the Greenland Ice Sheet with horizontal grid resolutions varying between 10 and 0.25 km. The model setup used as benchmark problem comprises a variety of modules with different levels of complexity and computational demands. The core builds the so-called stress balance module, which uses the higher-order approximation (or Blatter–Pattyn) of the Stokes equations, including free surface and ice-front evolution as well as thermodynamics in form of an enthalpy balance, and a mesh of linear prismatic finite elements, to compute the ice flow. We develop a detailed user-oriented, yet low-overhead, performance instrumentation tailored to the requirements of Earth system models and run scaling tests up to 6144 Message Passing Interface (MPI) processes. The results show that the computation of the Greenland model scales overall well up to 3072 MPI processes but is eventually slowed down by matrix assembly, the output handling and lower-dimensional problems that employ lower numbers of unknowns per MPI process. We also discuss improvements of the scaling and identify further improvements needed for climate research. The instrumented version of ISSM thus not only identifies potential performance bottlenecks that were not present at lower core counts but also provides the capability to continually monitor the performance of ISSM code basis. This is of long-term significance as the overall performance of ISSM model depends on the subtle interplay between algorithms, their implementation, underlying libraries, compilers, runtime systems and hardware characteristics, all of which are in a constant state of flux. We believe that future large-scale high-performance computing (HPC) systems will continue to employ the MPI-based programming paradigm on the road to exascale. Our scaling study pertains to a particular modelling setup available within ISSM and does not address accelerator techniques such as the use of vector units or GPUs. However, with 6144 MPI processes, we identified issues that need to be addressed in order to improve the ability of the ISSM code base to take advantage of upcoming systems that will require scaling to even higher numbers of MPI processes.

Detecting and Extracting Brain Hemorrhages from CT Images Using Generative Convolutional Imaging Scheme.

Purpose The need for computerized medical assistance for accurate detection of brain hemorrhage from Computer Tomography (CT) images is more mandatory than conventional clinical tests. Recent technologies and advanced computerized algorithms follow Artificial Intelligence (AI), Machine Learning (ML), and Deep Learning (DL) techniques to improve medical diagnosis platforms. This technology is making the diagnosis practice of brain issues easier for medical practitioners to analyze and identify diseases with an assured degree of precision and performance. Methods As the existing CT image analysis models use standard procedures to detect hemorrhages, the need for DL-based data analysis is essential to provide more accurate results. Generally, the existing techniques are limited with image training efficiency, image filtering procedures, and runtime system tuning modules. On the scope, this work develops a DL-based automated analysis of CT scan slices to find various levels of brain hemorrhages. Notably, this proposed system integrates Convolutional Neural Network (CNN) and Generative Adversarial Network (GAN) architectures as Integrated Generative Adversarial-Convolutional Imaging Model (IGACM) for extracting the CT image features for detecting brain hemorrhages. Results This system produces good results and takes lesser training time than existing techniques. This proposed system effectively works over CT images and classifies the abnormalities with more accuracy than current techniques. The experiments and results deliver the optimal detection of hemorrhages with better accuracy. It shows that the proposed system works with 5% to 10% of the better performance compared to other diagnostic techniques. Conclusion The complex nature of CT images leads to noncorrelated feature complexities in diagnosis models. Considering the issue, the proposed system used GAN-based effective sampling techniques for enriching complex image samples into CNN training phases. This concludes the effective contribution of the proposed IGACM technique for detecting brain hemorrhages than the existing diagnosis models.

Synthesis of Parallel Software from Heterogeneous Dataflow Models

Dataflow process networks (DPNs) are intrinsically data-driven, i.e., node actions are not synchronized among each other and may fire whenever sufficient input operands arrived at a node. While the general model of computation (MoC) of DPNs does not impose further restrictions, many different subclasses of DPNs representing different dataflow MoCs have been considered over time. These classes mainly differ in the kinds of behaviors of the processes. A DPN may be heterogeneous in that different processes in the network belong to different classes of DPNs. A heterogeneous DPN can therefore be effectively used to model and to implement different components of a system with different kinds of processes and, therefore, different dataflow MoCs. This paper presents a model-based design based on different dataflow MoCs including their heterogeneous combinations. In particular, it covers the automatic software synthesis of systems from DPN models. The main objective is to validate, evaluate and compare the artifacts exhibited by different dataflow MoCs at the implementation level of systems under the supervision of a common design tool. Moreover, this work also offers an efficient synthesis method that targets and exploits heterogeneity in DPNs by generating implementations based on the kinds of behaviors of the processes. The proposed synthesis method provides a tool chain including different specialized code generators for specific dataflow MoCs, and a runtime system that finally maps models using a combination of different dataflow MoCs on cross-vendor target hardware.

Run-Time Cutting Force Estimation Based on Learned Nonlinear Frequency Response Function

Abstract The frequency response function (FRF) provides an input–output model that describes the system dynamics. Learning the FRF of a mechanical system can facilitate system identification, adaptive control, and condition-based health monitoring. Traditionally, FRFs can be measured by off-line experimental testing, such as impulse response measurements via impact hammer testing. In this paper, we investigate learning FRFs from operational data with a nonlinear regression approach. A regression model with a learned nonlinear basis is proposed for FRF learning for run-time systems under dynamic steady state. Compared with a classic FRF, the data-driven model accounts for both transient and steady-state responses. With a nonlinear function basis, the FRF model naturally handles nonlinear frequency response analysis. The proposed method is tested and validated for dynamic cutting force estimation of machining spindles under various operating conditions. As shown in the results, instead of being a constant linear ratio, the learned FRF can represent different mapping relationships under different spindle speeds and force levels, which accounts for the nonlinear behavior of the systems. It is shown that the proposed method can predict dynamic cutting forces with high accuracy using measured vibration signals. We also demonstrate that the learned data-driven FRF can be easily applied with the few-shot learning scheme to machine tool spindles with different frequency responses when limited training samples are available.

Design of efficient single target real-time beam tracking system

This paper builds a system for real-time tracking of a single target Radio Frequency (RF) signal source, in which the transmitter uses a low-cost HackRF One software radio transceiver platform, and the receiving antenna array uses two monopole antennas to form a 1×2 receiving array. The receiver uses Xilinx 7-Series development board Zedboard and AD9361 RF transceiver to build a wireless receiving platform to process the received signal. In this RF beam tracking system, it needs to perform Direction of Arrival (DOA) estimation on the calibrated signal, the DOA algorithm plays a crucial role in the accuracy of beam pointing and system running time, the MUSIC algorithm takes some time to conduct spectrum peak search, so the Root-Music algorithm is used here to avoid the spectrum peak search problem. To enable the Root-MUSIC algorithm to be better executed on this two-element receiving platform, this work simplifies the entire calculation process, requiring only a small amount of calculation and running time to complete the DOA estimation step, which effectively save FPGA processing time and computing resources. It provides a guarantee for the real-time performance of the system.

Accelerating Geostatistical Modeling and Prediction With Mixed-Precision Computations: A High-Productivity Approach With PaRSEC

Geostatistical modeling, one of the prime motivating applications for exascale computing, is a technique for predicting desired quantities from geographically distributed data, based on statistical models and optimization of parameters. Spatial data are assumed to possess properties of stationarity or non-stationarity via a kernel fitted to a covariance matrix. A primary workhorse of stationary spatial statistics is Gaussian maximum log-likelihood estimation (MLE), whose central data structure is a dense, symmetric positive definite covariance matrix of the dimension of the number of correlated observations. Two essential operations in MLE are the application of the inverse and evaluation of the determinant of the covariance matrix. These can be rendered through the Cholesky decomposition and triangular solution. In this contribution, we reduce the precision of weakly correlated locations to single- or half- precision based on distance. We thus exploit mathematical structure to migrate MLE to a three-precision approximation that takes advantage of contemporary architectures offering BLAS3-like operations in a single instruction that are extremely fast for reduced precision. We illustrate application-expected accuracy worthy of double-precision from a majority half-precision computation, in a context where uniform single-precision is by itself insufficient. In tackling the complexity and imbalance caused by the mixing of three precisions, we deploy the <monospace>PaRSEC</monospace> runtime system. <monospace>PaRSEC</monospace> delivers on-demand casting of precisions while orchestrating tasks and data movement in a multi-GPU distributed-memory environment within a tile-based Cholesky factorization. Application-expected accuracy is maintained while achieving up to <inline-formula><tex-math notation="LaTeX">$1.59X$</tex-math></inline-formula> by mixing FP64/FP32 operations on 1536 nodes of <monospace>HAWK</monospace> or 4096 nodes of <monospace>Shaheen II</monospace> , and up to <inline-formula><tex-math notation="LaTeX">$2.64X$</tex-math></inline-formula> by mixing FP64/FP32/FP16 operations on 128 nodes of <monospace>Summit</monospace> , relative to FP64-only operations. This translates into up to 4.5, 4.7, and 9.1 (mixed) PFlop/s sustained performance, respectively, demonstrating a synergistic combination of exascale architecture, dynamic runtime software, and algorithmic adaptation applied to challenging environmental problems.

Design and Performance Characterization of RADICAL-Pilot on Leadership-Class Platforms

Many extreme scale scientific applications have workloads comprised of a large number of individual high-performance tasks. The Pilot abstraction decouples workload specification, resource management, and task execution via job placeholders and late-binding. As such, suitable implementations of the Pilot abstraction can support the collective execution of large number of tasks on supercomputers. We introduce RADICAL-Pilot (RP) as a portable, modular and extensible pilot-enabled runtime system. We describe RP's design, architecture and implementation. We characterize its performance and show its ability to scalably execute workloads comprised of tens of thousands heterogeneous tasks on DOE and NSF leadership-class HPC platforms. Specifically, we investigate RP's weak/strong scaling with CPU/GPU, single/multi core, (non)MPI tasks and Python functions when using most of ORNL Summit and TACC Frontera. RADICAL-Pilot can be used stand-alone, as well as the runtime for third-party workflow systems.

MAPPER: Managing Application Performance via Parallel Efficiency Regulation

State-of-the-art systems, whether in servers or desktops, provide ample computational and storage resources to allow multiple simultaneously executing potentially parallel applications. However, performance tends to be unpredictable, being a function of algorithmic design, resource allocation choices, and hardware resource limitations. In this article, we introduce MAPPER, a manager of application performance via parallel efficiency regulation. MAPPER uses a privileged daemon to monitor (using hardware performance counters) and coordinate all participating applications by making two coupled decisions: the degree of parallelism to allow each application to improve system efficiency while guaranteeing quality of service (QoS), and which specific CPU cores to schedule applications on. The QoS metric may be chosen by the application and could be in terms of execution time, throughput, or tail latency, relative to the maximum performance achievable on the machine. We demonstrate that using a normalized parallel efficiency metric allows comparison across and cooperation among applications to guarantee their required QoS. While MAPPER may be used without application or runtime modification, use of a simple interface to communicate application-level knowledge improves MAPPER’s efficacy. Using a QoS guarantee of 85% of the IPC achieved with a fair share of resources on the machine, MAPPER achieves up to 3.3 \( \times \) speedup relative to unmodified Linux and runtime systems, with an average improvement of 17% in our test cases. At the same time, MAPPER violates QoS for only 2% of the applications (compared to 23% for Linux), while placing much tighter bounds on the worst case. MAPPER relieves hardware bottlenecks via task-to-CPU placement and allocates more CPU contexts to applications that exhibit higher parallel efficiency while guaranteeing QoS, resulting in both individual application performance predictability and overall system efficiency.

PEM: Remote forensic acquisition of PLC memory in industrial control systems

Programmable logic controllers (PLC) are special-purpose embedded devices used in various industries for automatic control of physical processes. Cyberattacks on PLCs can unleash mayhem in the physical world. In case of a security breach, volatile memory acquisition is critical in investigating the attack since it provides unique insights into the runtime system activities and memory-based artifacts. However, existing memory acquisition methods for PLCs (i.e., using a hardware-level debugging port and network protocol-based approaches) are either inapplicable in real-world forensic investigations (due to requiring disassembling of a suspect PLC or power cycling) or incomplete (i.e., acquire only partial memory contents). This paper proposes a new memory acquisition framework to remotely acquire a PLC's volatile memory while the PLC is controlling a physical process. The main idea is to inject a harmless memory duplicator into the running control logic of a PLC to copy local memory contents into a protocol-mapped address space, which is then readable over a network. We also present a new control-logic attack that targets in-memory firmware to compromise a PLC's built-in system functions. Since PEM can acquire the entire PLC memory, we show that its memory dump contains evidence of this attack. Further, we present a case study on a gas pipeline testbed to demonstrate the effectiveness of the attack on a physical process and how PEM plays its role in effectively identifying the attack and other important forensic artifacts such as the control logic of a PLC.

Integration process simulator: A tool for performance evaluation of task scheduling of integration processes

ABSTRACT Due to large volumes of data from Cloud Computing and from the Internet of Things, the companies’ software ecosystem requires an efficient integration of applications and services. Performance improvement from integration platforms’ runtime systems is directly related to task scheduling strategies from integration processes. It is still a challenge to find the proper heuristic for a given integration process subject to high inbound data rates. This article proposes a simulation tool for the field of Enterprise Application Integration, which implements different scheduling heuristics and allows the extraction of performance metrics. Three task scheduling heuristics were simulated during the integration process, and the results were evaluated through statistical tests.

A lightweight approach to smart contracts supporting safety, security, and privacy

The concept of smart contract represents one of the most attractive uses of blockchain technology and has the advantage of being transparent, immutable, and corruption-free. However, blockchain is a highly resource demanding technology. The ambition of this paper is to propose a new approach for defining lightweight smart contracts, offering a high level of trust even without blockchain, when the underlying operating system can be trusted. Blockchain can be used for a higher degree of trust, for instance when the runtime system cannot be trusted. The approach gives transparency and immutability, and gives protection against corrupted or incorrect smart contract implementations. This is achieved by letting smart contract requirement specifications be separated from the smart contract implementations, provided by special objects, so-called history objects, recording all transactions of the associated contract. The history objects are generated by the runtime system as specially protected objects. Contract partners may interact with the history objects through predefined interfaces.We present a framework which includes an executable, imperative language for writing smart contracts, a functional language for contract specifications by means of invariants over the transaction history of a contract, as well as a verification system. The framework allows compositional and class-wise verification. A history object can provide runtime checking of specified behavioral properties of the contract, and can provide safety, security, and privacy control, as well as trusted transfer of assets. We demonstrate the approach on an auction system.