skip to main content

Title: The Price Is (Not) Right: Reflections on Pricing for Transient Cloud Servers
Amazon introduced spot instances in December 2009, enabling “customers to bid on unused Amazon EC2 capacity and run those instances for as long as their bid exceeds the current Spot Price.” Amazon’s real-time computational spot market was novel in multiple respects. For example, it was the first (and to date only) large-scale public implementation of market-based resource allocation based on dynamic pricing after decades of research, and it provided users with useful information, control knobs, and options for optimizing the cost of running cloud applications. Spot instances also introduced the concept of transient cloud servers derived from variable idle capacity that cloud platforms could revoke at any time. Transient servers have since become central to efficient resource management of modern clusters and clouds. As a result, Amazon’s spot market was the motivation for substantial research over the past decade. Yet, in November 2017, Amazon effectively ended its real-time spot market by announcing that users no longer needed to place bids and that spot prices will “...adjust more gradually, based on longer-term trends in supply and demand.” The changes made spot instances more similar to the fixed-price transient servers offered by other cloud platforms. Unfortunately, while these changes made spot instances more » less complex, they eliminated many benefits to sophisticated users in optimizing their applications. This paper provides a retrospective on Amazon’s real-time spot market, including its advantages and disadvantages for allocating transient servers compared to current fixed-price approaches. We also discuss some fundamental problems with Amazon’s spot market, which we identified in prior work (from 2016), that predicted its eventual end. We then discuss potential options for allocating transient servers that combine the advantages of Amazon’s real-time spot market, while also addressing the problems that likely led to its elimination. « less
; ; ; ; ;
Award ID(s):
Publication Date:
Journal Name:
2019 28th International Conference on Computer Communication and Networks (ICCCN)
Page Range or eLocation-ID:
1 to 9
Sponsoring Org:
National Science Foundation
More Like this
  1. Cloud platforms offer the same VMs under many purchasing options that specify different costs and time commitments, such as on-demand, reserved, sustained-use, scheduled reserve, transient, and spot block. In general, the stronger the commitment, i.e., longer and less flexible, the lower the price. However, longer and less flexible time commitments can increase cloud costs for users if future workloads cannot utilize the VMs they committed to buying. Large cloud customers often find it challenging to choose the right mix of purchasing options to reduce their long-term costs, while retaining the ability to adjust capacity up and down in response to workload variations.To address the problem, we design policies to optimize long-term cloud costs by selecting a mix of VM purchasing options based on short- and long-term expectations of workload utilization. We consider a batch trace spanning 4 years from a large shared cluster for a major state University system that includes 14k cores and 60 million job submissions, and evaluate how these jobs could be judiciously executed using cloud servers using our approach. Our results show that our policies incur a cost within 41% of an optimistic optimal offline approach, and 50% less than solely using on-demand VMs.
  2. Cloud users can significantly reduce their cost (by up to 60%) by reserving virtual machines (VMs) for long periods (1 or 3 years) rather than acquiring them on demand. Unfortunately, reserving VMs exposes users to demand risk that can increase cost if their expected future demand does not materialize. Since accurately forecasting demand over long periods is challenging, users often limit their use of reserved VMs. To mitigate demand risk, Amazon operates a Reserved Instance Marketplace (RIM) where users may publicly list the remaining time on their VM reservations for sale at a price they set. The RIM enables users to limit demand risk by either selling VM reservations if their demand changes, or purchasing variable- and shorter-term VM reservations that better match their demand forecast horizon. Clearly, the RIM’s potential to mitigate demand risk is a function of its price characteristics. However, to the best of our knowledge, historical RIM prices have neither been made publicly available nor analyzed. To address the problem, we have been monitoring and archiving RIM prices for 1.75 years across all 69 availability zones and 22 regions in Amazon’s Elastic Compute Cloud (EC2). This paper provides a first look at this data and itsmore »implications for cost-effectively provisioning cloud infrastructure.« less
  3. During the past few years, all leading cloud providers introduced burstable instances that can sprint their performance for a limited period to address sudden workload variations. Despite the availability of burstable instances, there is no clear understanding of how to minimize the waste of resources by regulating their burst capacity to the workload requirements. This is especially true when it comes to non-CPU-intensive applications. In this paper, we investigate how to limit network and I/O usage to optimize the efficiency of the bursting process. We also study which resource shall be controlled to benefit both cloud providers and end-users. We design MRburst (Multi-Resource burstable performance scheduler) to automatically limit multiple resources (i.e., network, I/O, and CPU) and make the application comply with a user-defined service level objective (SLO) while minimizing wasted resources. MRburst is evaluated on Amazon EC2 using two multi-resource applications: an FTP server and a Ceph system. Experimental results show that MRburst outperforms state-of-the-art approaches by allowing instances to speed up their performance for up to 2.4 times longer period while meeting SLO.
  4. Cloud providers offer instances with similar compute capabilities (for example, instances with different generations of GPUs like K80s, P100s, V100s) across many regions, availability zones, and on-demand and spot markets, with prices governed independently by individual supplies and demands. In this paper, using machine learning model training as an example application, we explore the potential cost reductions possible by leveraging this cross-cloud instance market. We present quantitative results on how the prices of cloud instances change with time, and how total costs can be decreased by considering this dynamic pricing market. Our preliminary experiments show that a) the optimal instance choice for a model is dependent on both the objective (e.g., cost, time, or combination) and the model’s performance characteristics, b) the cost of moving training jobs between instances is cheap, c) jobs do not need to be preempted more frequently than once a day to leverage the benefits from spot instance price variations, and d) the cost of training a model can be decreased by as much as 3.5× compared to a static policy. We also look at contexts where users specify higherlevel objectives over collections of jobs, show examples of policies for these contexts, and discuss additional challengesmore »involved in making these cost reductions viable.« less
  5. Problem definition: We seek to provide an interpretable framework for segmenting users in a population for personalized decision making. Methodology/results: We propose a general methodology, market segmentation trees (MSTs), for learning market segmentations explicitly driven by identifying differences in user response patterns. To demonstrate the versatility of our methodology, we design two new specialized MST algorithms: (i) choice model trees (CMTs), which can be used to predict a user’s choice amongst multiple options, and (ii) isotonic regression trees (IRTs), which can be used to solve the bid landscape forecasting problem. We provide a theoretical analysis of the asymptotic running times of our algorithmic methods, which validates their computational tractability on large data sets. We also provide a customizable, open-source code base for training MSTs in Python that uses several strategies for scalability, including parallel processing and warm starts. Finally, we assess the practical performance of MSTs on several synthetic and real-world data sets, showing that our method reliably finds market segmentations that accurately model response behavior. Managerial implications: The standard approach to conduct market segmentation for personalized decision making is to first perform market segmentation by clustering users according to similarities in their contextual features and then fit a “responsemore »model” to each segment to model how users respond to decisions. However, this approach may not be ideal if the contextual features prominent in distinguishing clusters are not key drivers of response behavior. Our approach addresses this issue by integrating market segmentation and response modeling, which consistently leads to improvements in response prediction accuracy, thereby aiding personalization. We find that such an integrated approach can be computationally tractable and effective even on large-scale data sets. Moreover, MSTs are interpretable because the market segments can easily be described by a decision tree and often require only a fraction of the number of market segments generated by traditional approaches. Disclaimer: This work was done prior to Ryan McNellis joining Amazon. Funding: This work was supported by the National Science Foundation [Grants CMMI-1763000 and CMMI-1944428]. Supplemental Material: The online appendices are available at .« less