With the increasing popularity of recommendation systems (RecSys), the demand
for compute resources in datacenters has surged. However, the model-wise
resource allocation employed in current RecSys model serving architectures
falls short in effectively utilizing resources, leading to sub-optimal total
cost of ownership. We propose ElasticRec, a model serving architecture for
RecSys providing resource elasticity and high memory efficiency. ElasticRec is
based on a microservice-based software architecture for fine-grained resource
allocation, tailored to the heterogeneous resource demands of RecSys.
Additionally, ElasticRec achieves high memory efficiency via our utility-based
resource allocation. Overall, ElasticRec achieves an average 3.3x reduction in
memory allocation size and 8.1x increase in memory utility, resulting in an
average 1.6x reduction in deployment cost compared to state-of-the-art RecSys
inference serving system.

ElasticRec 是一种模型服务架构，为推荐系统提供资源弹性性和高内存效率。相较于现有推荐系统预测服务系统，ElasticRec 通过微服务的软件架构和基于效用的资源分配实现了平均内存分配大小的 3.3 倍减少、内存效用的 8.1 倍增加，从而降低了 1.6 倍的部署成本。

ElasticRec：一种基于微服务的模型服务架构，实现对推荐模型的弹性资源扩缩容

ElasticRec: A Microservice-based Model Serving Architecture Enabling  Elastic Resource Scaling for Recommendation Models

We analyse the cost of using LLMs for planning and highlight that recent
trends are profoundly uneconomical. We propose a significantly more efficient
approach and argue for a responsible use of compute resources; urging research
community to investigate LLM-based approaches that upholds efficiency.

对使用 LLMs 进行规划的成本进行分析，指出最近的趋势极其不经济。我们提出了一种更高效的方法，并呼吁研究团体探索维持效率的基于 LLMs 的方法的负责任使用计算资源。

通过效率视角进行语言模型的规划

Planning with Language Models Through The Lens of Efficiency

Recently, the FourCastNet Neural Earth System Model (NESM) has shown
impressive results on predicting various atmospheric variables, trained on the
ERA5 reanalysis dataset. While FourCastNet enjoys quasi-linear time and memory
complexity in sequence length compared to quadratic complexity in vanilla
transformers, training FourCastNet on ERA5 from scratch still requires large
amount of compute resources, which is expensive or even inaccessible to most
researchers. In this work, we will show improved methods that can train
FourCastNet using only 1% of the compute required by the baseline, while
maintaining model performance or par or even better than the baseline.

在本研究中，我们将展示可以仅使用基准模型计算资源的 1% 来训练 FourCastNet，同时保持模型性能甚至比基准模型更好。