Memory vs. Efficiency ???

[ts4]

I have a program which has three main datastructures (linked lists):
1) Molecule [main linked list - a molecule file has several residues]
2) Res - linked list with it's own attributes
4) CloseRes - linked list (each Res has pointers to a structure which keeps track of the res close to this res)

There are two issues :
a) A molecule file (main input file) can have ~500 res [ so 500 node Res linked list]. If I have to process 10 such molecule files at the same time then what is the best way to do so ?
- Extending the molecule linked list is one way - but memory can be an issue here ???
- Writing output for each molecule to a file and then writing another program to process those files is another way -- efficiency seems to be an issue here???

b) For each set of "Close Res - pt 4 above" I have to do a computation for which I need to run a program considering the "CloseRes" and all the other molecule files --
- Should I just write each CloseRes to a file, run the program --repeat ---- not sure if this is the best approach

Given below is a code snippet representing the struct:
I will appreciate your suggestions. Thanks!

Code:

struct _res_type;
struct _closeres_type;
 
//res for a molecule
typedef struct _res_type
{
  char* resName;
  int resNo;
  float cx, cy, cz;
  float vScore;
  float sEnvScore;
 
  struct _closeres_type* closeres;
  struct _res_type* prev;
  struct _res_type* next;
}res_type;
 
//close_res for a res
typedef struct _closeres_type
{
  res_type* res;
  struct _closeres_type* next;
}closeres_type;
 
//molecule with the res
typedef struct _molecule_type
{
  int noRes;
  res_type* res;
  struct _molecule_type* prev;
  struct _molecule_type* next;
}molecule_type;
 
 

[weaknessforcats]

Are you using C or C++? The answer will be different.

[ts4]

I am using C for this. I will appreciate your response.
Thanks!

[weaknessforcats]

Your Molecule could be a linked list of 500 linked lists or maybe it is an array of 500 elements where each element is a pointer to a Res linked list.

This would be called a chain-link table. However, the linked list will work if the 500 figure is soft.

So you would have 10 of these tables.

Write a function to process one of these tables.

Write other functions to read/write these table to from/to disc. The read function would build the table in memory.

Your CloseRes would be handle the same way.

As to memory, do a sizeof on all of your structs and multiply out the total. I expect everything will fit in memory.

I see a program that does data acquisition, data processing, and data store/retrieval. Unless you are dealing with hundreds of thousands of lines of code, then probably one program will be the ticket.

[ts4]

Thanks for your response. I did extend my molecule linked list as suggested by you. So, now my molecule linked list has 10(current max) molecule nodes and each molecule node is linked to another linked list with 100-500 nodes.

It works for 100 nodes but I haven't yet tested it for 500 nodes.
Thanks for your help!