diff --git a/README b/README.md
similarity index 84%
rename from README
rename to README.md
index 547195d..96a2cf8 100644
--- a/README
+++ b/README.md
@@ -1,4 +1,19 @@
-Note: In our research paper we use the SSB instead of SSBM
+Star Schema Benchmark data generator
+==
+
+This branch has been updated to fix building of the utility on Linux.
+
+Original problems with the utility remain, such as crashes.  
+Generate each table individually:
+* ./dbgen -s 10 -T s 
+* ./dbgen -s 10 -T d
+* ./dbgen -s 10 -T p 
+* ./dbgen -s 10 -T c
+* ./dbgen -s 10 -T l
+
+Original Readme
+--
+Note: In ~~our research paper we~~~ the original paper the authors use the SSB instead of SSBM
 Version of 2/28/10: 
 Cardinality of supplier fixed to follow benchmark spec: now 2000*SF
  (previously was 10000*SF, in error): line 226, driver.c
diff --git a/bm_utils.c b/bm_utils.c
index 5da29a0..6194909 100644
--- a/bm_utils.c
+++ b/bm_utils.c
@@ -374,12 +374,12 @@ tbl_open(int tbl, char *mode)
         fprintf(stderr, "stat(%s) failed.\n", fullpath);
         exit(-1);
         }
-    if (S_ISREG(fstats.st_mode) && !force && *mode != 'r' )
+    /*if (S_ISREG(fstats.st_mode) && !force && *mode != 'r' )
         {
-        sprintf(prompt, "Do you want to overwrite %s ?", fullpath);
+         sprintf(prompt, "Do you want to overwrite %s ?", fullpath);
         if (!yes_no(prompt))
             exit(0);
-        }
+        } */
 
     if (S_ISFIFO(fstats.st_mode))
         {
@@ -389,17 +389,7 @@ tbl_open(int tbl, char *mode)
         }
     else{
 
-#ifdef LINUX
-      /* allow large files on Linux */
-      /*use open to first to get the in fd and apply regular fdopen*/
-
-	/*cheng: Betty mentioned about write mode problem here, added 066*/
-      retcode =
-		  open(fullpath, ((*mode == 'r')?O_RDONLY:O_WRONLY)|O_CREAT|O_LARGEFILE,0644);
-        f = fdopen(retcode, mode);
-#else
         f = fopen(fullpath, mode);
-#endif
 
     }
     OPEN_CHECK(f, fullpath);
diff --git a/config.h b/config.h
index 91833cb..dbdffe0 100644
--- a/config.h
+++ b/config.h
@@ -47,6 +47,9 @@
  *   SYBASE     -- use Sybase dialect in QGEN
  *   TDAT       -- use Teradata dialect in QGEN
  */
+#ifndef pid_t
+#include <sys/types.h>
+#endif
 
 #ifdef DOS
 #define DSS_PROC        1
diff --git a/gendata b/gendata
new file mode 100755
index 0000000..3370137
--- /dev/null
+++ b/gendata
@@ -0,0 +1,27 @@
+#!/bin/bash
+set -e
+if [ "$1" = "" ]; then
+  echo "No scale factor provided.  Using scale factor 1."
+  echo "To use another scale factor use $0 X where X is an integer value between 1 and 1000"
+  SIZE=1
+else
+  SIZE=$1
+fi;
+if [ -e lineorder.tbl ]; then
+  echo "Backing up existing data set"
+  TS=$(date '+%s')
+  BACKUPDIR="backup.$TS"
+  mkdir $BACKUPDIR
+  mv *.tbl $BACKUPDIR
+  echo "Backup completed"
+fi;
+echo "Generating new data set of scale factor $SIZE"  
+./dbgen -s $SIZE -T l > /dev/null 2>&1 &
+./dbgen -s $SIZE -T c > /dev/null 2>&1 &
+./dbgen -s $SIZE -T d > /dev/null 2>&1 &
+./dbgen -s $SIZE -T p > /dev/null 2>&1 &
+./dbgen -s $SIZE -T s > /dev/null 2>&1 &
+wait
+echo "Data generation completed!"
+du -sh *.tbl
+
diff --git a/shared.h b/shared.h
index c1c18ce..606f062 100644
--- a/shared.h
+++ b/shared.h
@@ -117,7 +117,7 @@
 #define  PHONE_LEN     15
 
 #ifdef SSBM
-#define  MAXAGG_LEN    10    /* max component length for a agg str */
+#define  MAXAGG_LEN    20    /* max component length for a agg str */
 
 #else
 #define  MAXAGG_LEN    20    /* max component length for a agg str */