Cloud computing offers the potential to store, manage, and process data in highly available, scalable, and elastic environments. Yet, these environments still provide very limited and inflexible means for customers to control their data. For example, customers can neither specify security of inter-cloud communication bearing the risk of information leakage, nor comply with laws requiring data to be kept in the originating jurisdiction, nor control sharing of data with third parties on a fine-granular basis. This lack of control can hinder cloud adoption for data that falls under regulations. In this paper, we show in six use cases how cloud environments can be enriched with policy language support to give customers control over cloud data. Our use cases are based on realizing policy language support in all three cloud environment layers, i.e., IaaS, PaaS, and SaaS. Specifically, we present policy-aware resource management (with OpenStack) and dynamic network configuration. With CERN's big data storage and the in-memory database Hyrise, we show realization for storage and further exemplify policy-aware cloud processing by network function virtualization which enables Orange to offload customer home gateways to the cloud. Finally, we discuss benefits of policy support in F-Secure's Security Cloud. These use cases show the feasibility of realizing customer control with policy support in the cloud. Thus, our work enables customers with regulated data to tap cloud benefits and significantly broadens the market for cloud providers.

The growing analytical demand increases the importance of scalability and elasticity for mixed workload in-memory databases. Data replication is a way to cope with the growing demand and entails increased availability. In this paper, we describe different replication mechanisms, balancing query performance and availability. In addition, we outline how we implemented the cloud-ilities scalability, availability, and elasticity in Hyrise-R, a replication extension of the in-memory database Hyrise. Finally, we summarize further current research activities within the Hyrise project, i. e., data tiering, self-adaption and non-volatile RAM.

While single machines can handle the transactional database workload of most companies, the increasing analytical load will push them to their limit. For this reason, we extended the open source in-memory database Hyrise with the capability to form a database cluster for scalability and increased availability. This scale out and hot standby version is called Hyrise-R. It implements lazy master replication and has been shown to be well suited for mixed workloads as they exist in enterprise applications. In this paper we present our extension of Hyrise-R: a query dispatcher, which works fully transparently and implements an enhanced query distribution algorithm. The new distribution algorithm improves load balancing and prioritizes write requests for higher transaction throughput. In addition, we discuss our work in progress and planned activities for Hyrise-R.

One well established method of measuring the success of companies are key performance indicators, whose inter-dependencies can be represented by mathematical models, such as value driver trees. While such models have commonly agreed semantics, they lack the right tool support for business simulations, because a flexible implementation that supports multi-dimensional and hierarchical structures on large data sets is complex and computationally challenging. However, in-memory column stores as the backbone of enterprise applications provide incredible performance that enables to calculate flexible simulation scenarios interactively even on large sets of enterprise data.

In-memory database systems are well-suited for enterprise workloads, consisting of transactional and analytical queries. A growing number of users and an increasing demand for enterprise applications can saturate or even overload single- node database systems at peak times. Better performance can be achieved by improving a single machine’s hardware but it is often cheaper and more practicable to follow a scale-out approach and replicate data by using additional machines. In this paper we present Hyrise-R, a lazy master replication system for the in-memory database Hyrise. By setting up a snapshot-based Hyrise cluster, we increase both performance by distributing queries over multiple instances and availability by utilizing the redundancy of the cluster structure. This paper describes the architecture of Hyrise- R and details of the implemented replication mechanisms. We set up Hyrise-R on instances of Amazon’s Elastic Compute Cloud and present a detailed performance evaluation of our system, including a linear query throughput increase for enterprise workloads.

Value driver trees are a well-known methodology to model dependencies such as the definition of key performance indicators. While the models have well-known semantics, they lack the right tool support for business simulations, because a flexible implementation that supports multidimensional, hierarchical value driver trees and data bindings is very complex and computationally challenging. This paper tackles this problem by proposing an approach for generic enterprise simulations which are based on value driver trees. Our approach is two-fold: we present the definition of a simulation meta model at design time, and the run-time simulation tool. The simulation meta model describes the structure of the dependency graph, the data binding, and the parametrization of the model to simulate data changes. The simulation tool can then be used to create and edit simulation model instances and run simulations in real-time by leveraging an in-memory column store. Besides the formal description of the approach, this work presents a prototypical implementation of the simulation tool and an evaluation using data of a consumer packaged goods company.

Available-To-Promise (ATP) is an application in the context of Supply Chain Management (SCM) systems and provides a checking mechanism that calculates if the desired products of a customer order can be delivered on the requested date. Modern SCM systems store relevant data records as aggregated numbers which implies the disadvantages of maintaining redundant data as well as inflexibility in querying the data. Our approach omits aggregates by storing all individual data records in an in-memory, column-store and scans through all relevant records on-the-fly for each check. We contribute by describing the novel data organization and a lockingfree, highly-concurrent ATP checking algorithm. Additionally, we explain how new business functionality such as instant rescheduling of orders can be realized with our approach. All concepts are implemented within a prototype and benchmarked by using an anonymized SCM dataset of a Fortune 500 consumer products company. The paper closes with a discussion of the results and gives an outlook how this approach can help companies to find the right balance between low inventory costs and high order fulfillment rates.