Search

Software-defined storage (SDS) is an abrupt technology that aims to address the explosive data proliferation and storage management complexity through the separation of control and data path, which are tightly bounded in the traditional storage area network (SAN) model. In SDS, the centralized controller manages and controls the overall storage services and operations. However, the centralized controller approach raises the scalability and single point of failure flag. In such a paradigm, the controller placement is an important activity. A random placement of controllers suffers from unpredictable delay in a large enterprise. Multiple SDS controllers mitigate the single point of failure and provide redundancy. We propose a method called controller placement based on the center of gravity (CPCG) to solve the optimal placement controllers’ locations to reduce the delay. CPCG incorporates different coordinates and analyzes their density (the number of attached devices to a switch) to suggest new optimal sites. For the least number of SDS controllers, we propose an areawise optimal placement location strategy; the areawise approach mitigates workload processing from the central controller and also provides a scalable and reliable storage system. Experiments show that the optimal placement drastically reduces delay and, hence, boosts scalability.

In extrusion of hollow Al-profiles two kinds of pressure welds are present inside the extrusion. One is called the charge weld (CW) and forms across the boundary interface between two billets extruded in sequence. The other is the seam weld (SW) which extends longitudinally along the extruded profile and the extrusion metal behind each die bridge. It is considered to form because of the splitting of the extrusion metal over the die bridge into metal streams which flow past the bridge and rejoin as they encounter behind the bridge. Over the time attempts have been made to explain the mechanics of extrusion welding for both the CW and the SW. Still there is lack of understanding of how these welds form, the main reasons for this is that the deformation conditions around a die bridge are complex and difficult to investigate. Because of the recent advancement of two technological fields, experimental grid pattern analysis and simulation of metal flow by FEA; new tools for analysis of the mechanics of formation of the SW and the CW are now available. The simplest possible case of 2D-extrusion seam welding is considered here and an attempt is made to describe the fundamental deformation mechanisms present when this weld forms behind a butt-ended die bridge.

In extrusion of aluminium alloys, using porthole dies, the billet metal is split into separate metal streams during flow into the ports. After that, the streams join by seam pressure welding in the weld chamber of the die and finally flow through the die exit emerging as the desired extrusion. Because of re-joining of the metal streams the extrusion will contain as many extrusion seam welds as there are bridges in the die. Dependent on the die design the metal will flow in such a way that a gas pocket may form behind the die bridge and remain stable here throughout the course of the extrusion process. If, at this stage of the process, the feed of metal into the region behind the bridge is increased the space behind the bridge may fill up completely and the gas pocket will disappear. In this work we have investigated how finite element modeling can be used (for a simple idealized case of 2-D extrusion welding) to understand the extrusion seam welding process, and to characterize what extrusion conditions will provide stable gas pocket formation behind a bridge, and when the pocket will disappear. The geometry of the welding chamber and the die opening was varied in a number of simulations to study this phenomenon using the FEM-code Deform 2D®. Based on these simulations we have been able to propose an “extrusion seam weld limit diagram (ESWLD)”. The ESWLD shows when extrusion will occur with a gas pocket present behind the bridge and when there is transfer to a state where the pocket disappears and the die fills up completely with metal. By curve-fitting the results can be presented by a mathematical model, i.e. by an equation allowing the reduction in extrusion at the point of transfer from a state of presence of the pocket to a state of complete die filling, to be computed from the dimensions of the porthole and the weld chamber, for a given billet size and a given bridge geometry. The model also shows which geometrical extrusion parameters of the die will favor complete filling of the space behind a die bridge instead of gas pocket formation there.

Hollow and semi-hollow profiles are commonly produced by extrusion using porthole dies. The main characteristics of such dies are the presence of a mandrel (core) to shape the inner contour of hollow profile and bridges or legs to carry the mandrel. The bridges split the billet material into multiple metal streams that flow through the porthole channels and meet in the welding chamber behind the bridge where they are joined by pressure welding. When hollow profiles with different wall thickness are made the size of two adjacent portholes may be different. The material then flows through the two portholes with different flow velocity so that there is more feed through the bigger porthole into the weld chamber behind the bridge. Experiments have been performed and are reported here in which a grid pattern technique was used to characterize the metal flow through a 2D-die with porthole channels of unequal size. The design of the laboratory die has been modified in relation to the symmetric case to get different sizes of the two portholes. Since the metal flow through such a die is asymmetric the grid pattern technique was also modified to characterize the experimental flow. The results of an experimental metal flow study performed for a short billet was presented in a previous article [1]. Corresponding experiments performed with longer billets are now reported; so that two stages of the extrusion process is analysed here. The grid pattern technique has successfully mapped the non-symmetric material flow as in industrial extrusion when using different wall thickness over the section. The lateral movement of metal during extrusion is obtained from one set of experiments; the vertical movement from the other set. Finite element analysis of the extrusion process has been performed using Deform 3D. The encountering of the two metal streams behind the die bridge and the deformation characteristics within the welding chamber has been studied this way. Extrusion weld formation and deformations around the die bridge are considered here with the help of experimental results and simulation models. The nature of the metal flow achieved from the FE-model is compared with the experimental results. As regards the short billet some results are presented in [1], however improvement to the previous model gives a more perfect match. The model also provides information about the boundary conditions in real extrusion.

The article presents an outline of a scientific approach for testing constitutive relations for the aluminum extrusion process. By comparing ram force, container friction, die face pressure, outlet temperature measurement during rod extrusion with corresponding simulated data, inferences can in principle be drawn with respect to the validity models. The paper indicates that simulation results from the 2D ALMA2π program are in fair agreement with measurements during extrusion of AA6060, but more work needs to be done to control thermal conditions during extrusion.